Researchers have introduced RAGEN, an AI framework designed to counter LLM agent instability when handling complex situations. Training these AI agents presents significant hurdles, particularly when decisions span multiple steps and involve unpredictable feedback from the environment. While reinforcement learning (RL) has shown promise in static tasks like solving maths problems or generating code, its application to dynamic, multi-turn agent training has been less explored.    Addressing this gap, a collaborative team from institutions including Northwestern University, Stanford University, Microsoft, and New York University has proposed StarPO (State-Thinking-Actions-Reward Policy Optimisation). StarPO offers a generalised approach for training agents at the trajectory level (i.e. it optimises the entire sequence of interactions, not just individual actions.) Accompanying this is RAGEN, a modular system built to implement StarPO. This enables the training and evaluation of LLM agents, particularly focusing on their reasoning capabilities under RL. RAGEN provides the necessary infrastructure for rollouts, reward assignment, and optimisation within multi-turn, stochastic (randomly determined) environments. Minimalist environments, maximum insight To isolate the core learning challenges from confounding factors like extensive pre-existing knowledge or task-specific engineering, the researchers tested LLMs using RAGEN in three deliberately minimalistic, controllable symbolic gaming environments:    Bandit: A single-turn, stochastic task testing risk-sensitive symbolic reasoning. The agent chooses between options (like ‘Phoenix’ or ‘Dragon’ arms) with different, initially unknown, reward profiles. Sokoban: A multi-turn, deterministic puzzle requiring foresight and planning, as actions (pushing boxes) are irreversible. Frozen Lake: A multi-turn, stochastic grid navigation task where movement attempts can randomly fail, demanding planning under uncertainty. These environments allow for clear analysis of how agents learn decision-making policies purely through interaction.    Key findings: Stability, rollouts, and reasoning The study yielded three significant findings concerning the training of self-evolving LLM agents: The ‘Echo Trap’ and the need for stability A recurring problem observed during multi-turn RL training was dubbed the “Echo Trap”. Agents would initially improve but then suffer performance collapse, overfitting to locally rewarded reasoning patterns.  This was marked by collapsing reward variance, falling entropy (a measure of randomness/exploration), and sudden spikes in gradients (indicating training instability). Early signs included drops in reward standard deviation and output entropy.    To combat this, the team developed StarPO-S, a stabilised version of the framework. StarPO-S incorporates:    Variance-based trajectory filtering: Focusing training on task instances where the agent’s behaviour shows higher uncertainty (higher reward variance), discarding low-variance, less informative rollouts. This improved stability and efficiency.    Critic incorporation: Using methods like PPO (Proximal Policy Optimisation), which employ a ‘critic’ to estimate value, generally showed better stability than critic-free methods like GRPO (Group Relative Policy Optimisation) in most tests.    Decoupled clipping and KL removal: Techniques adapted from other research (DAPO) involving asymmetric clipping (allowing more aggressive learning from positive rewards) and removing KL divergence penalties (encouraging exploration) further boosted stability and performance.    StarPO-S consistently delayed collapse and improved final task performance compared to vanilla StarPO.    Rollout quality is crucial The characteristics of the ‘rollouts’ (simulated interaction trajectories used for training) significantly impact learning. Key factors identified include:    Task diversity: Training with a diverse set of initial states (prompts), but with multiple responses generated per prompt, aids generalisation. The sweet spot seemed to be moderate diversity enabling contrast between different outcomes in similar scenarios.    Interaction granularity: Allowing multiple actions per turn (around 5-6 proved optimal) enables better planning within a fixed turn limit, without introducing the noise associated with excessively long action sequences.    Rollout frequency: Using fresh, up-to-date rollouts that reflect the agent’s current policy is vital. More frequent sampling (approaching an ‘online’ setting) leads to faster convergence and better generalisation by reducing policy-data mismatch. Maintaining freshness, alongside appropriate action budgets and task diversity, is key for stable training.    Reasoning requires careful reward design Simply prompting models to ‘think’ doesn’t guarantee meaningful reasoning emerges, especially in multi-turn tasks. The study found: Reasoning traces helped generalisation in the simpler, single-turn Bandit task, even when symbolic cues conflicted with rewards.    In multi-turn tasks like Sokoban, reasoning benefits were limited, and the length of ‘thinking’ segments consistently declined during training. Agents often regressed to direct action selection or produced “hallucinated reasoning” if rewards only tracked task success, revealing a “mismatch between thoughts and environment states.” This suggests that standard trajectory-level rewards (often sparse and outcome-based) are insufficient.  “Without fine-grained, reasoning-aware reward signals, agent reasoning hardly emerge[s] through multi-turn RL.” The researchers propose that future work should explore rewards that explicitly evaluate the quality of intermediate reasoning steps, perhaps using format-based penalties or rewarding explanation quality, rather than just final outcomes.    RAGEN and StarPO: A step towards self-evolving AI The RAGEN system and StarPO framework represent a step towards training LLM agents that can reason and adapt through interaction in complex, unpredictable environments. This research highlights the unique stability challenges posed by multi-turn RL and offers concrete strategies – like StarPO-S’s filtering and stabilisation techniques – to mitigate them. It also underscores the critical role of rollout generation strategies and the need for more sophisticated reward mechanisms to cultivate genuine reasoning, rather than superficial strategies or hallucinations. While acknowledging limitations – including the need to test on larger models and optimise for domains without easily verifiable rewards – the work opens “a scalable and principled path for building AI systems” in areas demanding complex interaction and verifiable outcomes, such as theorem proving, software engineering, and scientific discovery. (Image by Gerd Altmann) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

There is a sense of urgency in Europe to re-imagine the status quo and reshape technology infrastructures. Timed to harness Europe’s innovative push comes GITEX EUROPE x Ai Everything (21-23 May, Messe Berlin). The world’s third largest economy and host nation for GITEX EUROPE x Ai Everything, Germany’s role as the European economic and technology leader is confirmed as its ICT sector is projected to reach €232.8bn in 2025 (Statista). GITEX EUROPE x Ai Everything is Europe’s largest tech, startup and digital investment event, and is organised by KAOUN International. It’s hosted in partnership with the Berlin Senate Department for Economics, Energy and Public Enterprises, Germany’s Federal Ministry for Economic Affairs and Climate Action, Berlin Partner for Business and Technology, and the European Innovation Council (EIC). Global tech engages for cross-border and industry partnerships The first GITEX EUROPE brings together over 1,400 tech enterprises, startups and SMEs, and platinum sponsors AWS and IBM. Also in sponsorship roles are Cisco, Cloudflare, Dell, Fortinet, Lenovo, NTT, Nutanix, Nvidia, Opswat, and SAP. GITEX EUROPE x Ai Everything will comprise of tech companies from over 100 countries and 34 European states, including tech pavilions from India, Italy, Morocco, Netherlands, Poland, Serbia, South Korea, UK, and the UAE. Trixie LohMirmand, CEO of KAOUN International, organiser of GITEX worldwide, said: “There is a sense of urgency and unity in Europe to assert its digital sovereignty and leadership as a global innovation force. The region is paving its way as a centre-stage where AI, quantum and deep tech will be debated, developed, and scaled.” Global leaders address EU’s tech crossroads Organisers state there will be over 500 speakers, debating a range of issues including AI and quantum, cloud, and data sovereignty. Already confirmed are Geoffrey Hinton, Physics Nobel Laureate (2024); Kai Wegner, Mayor of Berlin; H.E. Jelena Begović, Serbian Minister of Science, Technological Development and Innovation; António Henriques, CEO, Bison Bank; Jager McConnell, CEO, Crunchbase; Mark Surman, President, Mozilla; and Sandro Gianella, Head of Europe & Middle East Policy & Partnerships, OpenAI. Europe’s moves in AI, deep tech & quantum Europe is focusing on cross-sector AI uses, new investments and international partnerships. Ai Everything Europe, the event’s AI showcase and conference, brings together AI architects, startups and investors to explore AI ecosystems. Topics presented on stage range from EuroStack ambitions to implications of agentic AI, with speakers including Martin Kon, President and COO, Cohere; Daniel Verten, Strategy Partner, Synthesia; and Professor Dr. Antonio Krueger, CEO of German Research Centre for Artificial Intelligence. On the show-floor, attendees will be able to experience Brazil’s Ubivis’s smart factory technology, powered by IoT and digital twins, and Hexis’s AI-driven nutrition plans that are trusted by 500+ Olympic and elite athletes. With nearly €7 billion in quantum investment, Europe is pushing for quantum leadership by 2030. GITEX Quantum Expo (GQX) (in partnership with IBM and QuIC) covers quantum research and cross-industry impact with showcases and conferences. Speakers include Mira Wolf-Bauwens, Responsible Quantum Computing Lead, IBM Research, Switzerland; Joachim Mnich, Director of Research & Computing, CERN, Switzerland; Neil Abroug, Head of the French National Quantum Strategy, INRIA; and Jan Goetz, CEO & Co-Founder, IQM Quantum Computers, Finland. Cyber Valley: Building a resilient cyber frontline With cloud breaches doubling in number and AI-driven attacks, threat response and cyber resilience are core focuses at the event. Fortinet, CrowdStrike, Kaspersky, Knowbe4, and Proofpoint will join other cybersecurity companies exhibiting at GITEX Cyber Valley. They’ll be alongside law enforcement leaders, global CISOs, and policymakers on stage, including Brig. Gen. Dr. Volker Pötzsch, Chief of Division Cyber/IT & AI, Federal Ministry of Defence, Germany; H.E. Dr. Mohamed Al-Kuwaiti, Head of Cybersecurity, UAE Government; Miguel De Bruycker, Managing Director General, Centre for Cybersecurity Belgium; and Ugo Vignolo Lutati, Group CISO, Prada Group. GITEX Green Impact: For a sustainable future GITEX Green Impact connects innovators and investors with over 100 startups and investors exploring how green hydrogen, bio-energy, and next-gen energy storage are moving from R&D to deployment. Key speakers so far confirmed are Gavin Towler, Chief Scientist for Sustainability Technologies & CTO, Honeywell; Julie Kitcher, Chief Sustainability Officer, Airbus; Lisa Reehten, Managing Director, Bosch Climate Solutions; Massimo Falcioni, Chief Competitiveness Officer, Abu Dhabi Investment Office; and Mounir Benaija, CTO – EV & Charging Infrastructure, TotalEnergies. Convening the largest startup ecosystem among 60+ nations GITEX EUROPE x Ai Everything hosts North Star Europe, the local version of the world’s largest startup event, Expand North Star. North Star Europe gathers over 750 startups and 20 global unicorns, among them reMarkable, TransferMate, Solarisbank AG, Bolt, Flix, and Glovo. The event features a curated collection of earlys and growth-stage startups from Belgium, France, Hungary, Italy, Morocco, Portugal, Netherlands, Switzerland, Serbia, UK, and UAE. Among the startups, Neurocast.ai (Netherlands) is advancing AI-powered neurotech for Alzheimer’s research; CloudBees (Switzerland) is the delivery unicorn backed by Goldman Sachs, HSBC, and Lightspeed; and Semiqon (Finland), the world’s first CMOS transistor with the ability to perform in cryogenic conditions. More than 600 investors with $1tn assets under management will be scouting for new opportunities, including Germany’s Earlybird VC, Austria’s SpeedInvest, Switzerland’s B2Venture, Estonia’s Startup Wise Guys, and the US’s SOSV. GITEX ScaleX launches as a first-of-its-kind growth platform for scale-ups and late-stage companies, in partnership with AWS. With SMEs making up 99% of European businesses, GITEX SMEDEX connects SMEs with international trade networks and investors, for funding, legal advice, and market access to scale globally. Backed by EISMEA and ICC Digital Standards Initiative, the event features SME ecosystem leaders advising from the stage, including Milena Stoycheva, Chairperson of Board of Innovation, Ministry of Innovation and Growth, Bulgaria; and Oliver Grün, President, European Digital SME Alliance and BITMi. GITEX EUROPE is part of the GITEX global network tech and startup events, taking place in Germany, Morocco, Nigeria, Singapore, Thailand, and the UAE. For more information, please visit: www.gitex-europe.com.

AI memory demand has catapulted SK Hynix to a top position in the global DRAM market, overtaking longtime leader Samsung for the first time. According to Counterpoint Research data, SK Hynix captured 36% of the DRAM market in Q1 2025, compared to Samsung’s 34% share. HBM chips drive market shift The company’s achievement ends Samsung’s three-decade dominance in DRAM manufacturing and comes shortly after SK Hynix’s operating profit passed Samsung’s in Q4 2024. The company’s strategic focus on high-bandwidth memory (HBM) chips, essential components for artificial intelligence applications, has proven to be the decisive factor in the market shift. “The is a milestone for SK Hynix which is successfully delivering on DRAM to a market that continues to show unfettered demand for HBM memory,” said Jeongku Choi, senior analyst at Counterpoint Research. “The manufacturing of specialised HBM DRAM chips has been notoriously tricky and those that got it right early on have reaped dividends.” SK Hynix has taken the overall DRAM market lead and has established its dominance in the HBM sector, occupying 70% of this high-value market segment, according to Counterpoint Research. HBM chips, which stack multiple DRAM dies to dramatically increase data processing capabilities, have become fundamental components for training AI models. “It’s another wake-up call for Samsung,” said MS Hwang, research director at Counterpoint Research in Seoul, as quoted by Bloomberg. Hwang noted that SK Hynix’s leadership in HBM chips likely comprised a larger portion of the company’s operating income. Financial performance and industry outlook The company is expected to report positive financial results on Thursday, with analysts projecting a 38% quarterly rise in sales and a 129% increase in operating profit for the March quarter, according to Bloomberg data. The shift in market leadership reflects broader changes in the semiconductor industry as AI applications drive demand for specialised memory solutions. While traditional DRAM remains essential for computing devices, HBM chips that can handle the enormous data requirements of generative AI systems are becoming increasingly valuable. Market research firm TrendForce forecasts that SK Hynix will maintain its leadership position throughout 2025, coming to control over 50% of the HBM market in gigabit shipments. Samsung’s share is expected to decline to under 30%, while Micron Technology is said to gain ground to take close to 20% of the market. Counterpoint Research expects the overall DRAM market in Q2 2025 to maintain similar patterns across segment growth and vendor share, suggesting SK Hynix’s newfound leadership position may be sustainable in the near term. Navigating potential AI memory demand headwinds Despite the current AI memory demand boom, industry analysts identify several challenges on the horizon. “Right now the world is focused on the impact of tariffs, so the question is: what’s going to happen with HBM DRAM?” said MS Hwang. “At least in the short term, the segment is less likely to be affected by any trade shock as AI demand should remain strong. More significantly, the end product for HBM is AI servers, which – by definition – can be borderless.” However, longer-term risks remain significant. Counterpoint Research sees potential threats to HBM DRAM market growth “stemming from structural challenges brought on by trade shock that could trigger a recession or even a depression.” Morgan Stanley analysts, led by Shawn Kim, expressed similar sentiment in a note to investors cited by Bloomberg: “The real tariff impact on memory resembles an iceberg, with most danger unseen below the surface and still approaching.” The analysts cautioned that earnings reports might be overshadowed by these larger macroeconomic forces. Interestingly, despite SK Hynix’s current advantage, Morgan Stanley still favours Samsung as their top pick in the memory sector. “It can better withstand a macro slowdown, is priced at trough multiples, has optionality of future growth via HBM, and is buying back shares every day,” analysts wrote. Samsung is scheduled to provide its complete financial statement with net income and divisional breakdowns on April 30, after reporting preliminary operating profit of 6.6 trillion won ($6 billion) on revenue of 79 trillion won earlier this month. The shift in competitive positioning between the two South Korean memory giants underscores how specialised AI components are reshaping the semiconductor industry. SK Hynix’s early and aggressive investment in HBM technology has paid off, though Samsung’s considerable resources ensure the rivalry will continue. For the broader technology ecosystem, the change in DRAM market leadership signals the growing importance of AI-specific hardware components. As data centres worldwide continue expanding to support increasingly-sophisticated AI models, AI memory demand should remain robust despite potential macroeconomic headwinds. (Image credit: SK Hynix) See also: Samsung aims to boost on-device AI with LPDDR5X DRAM Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here. The post AI memory demand propels SK Hynix to historic DRAM market leadership appeared first on AI News.

China’s tech companies will drive adoption of the MCP (Model Context Protocol) standard that transforms AI assistants from simple chatbots into powerful digital helpers. MCP works like a universal connector that lets AI assistants interact directly with favourite apps and services – enabling them to make payments, book appointments, check maps, and access information on different platforms on users’ behalves. As reported by the South China Morning Post, companies like Ant Group, Alibaba Cloud, and Baidu are deploying MCP-based services and positioning AI agents as the next step, after chatbots and large language models. But will China’s MCP adoption truly transform the AI landscape, or is it simply another step in the technology’s evolution? Why China’s MCP adoption matters for AI’s evolution The Model Context Protocol was initially introduced by Anthropic in November 2024, at the time described as a standard that connects AI agents “to the systems where data lives, including content repositories, business tools and development environments.” MCP serves as what Ant Group calls a “USB-C port for AI applications” – a universal connector allowing AI agents to integrate with multiple systems. The standardisation is particularly significant for AI agents like Butterfly Effect’s Manus, which are designed to autonomously perform tasks by creating plans consisting of specific subtasks using available resources. Unlike traditional chatbots that just respond to queries, AI agents can actively interact with different systems, collect feedback, and incorporate that feedback into new actions. China’s MCP adoption by tech leaders highlights the importance placed on AI agents as the next evolution in artificial intelligence: Ant Group, Alibaba’s fintech affiliate, has unveiled its “MCP server for payment services,” that lets AI agents connect with Alipay’s payment platform. The integration allows users to “easily make payments, check payment statuses and initiate refunds using simple natural language commands,” according to Ant Group’s statement. Additionally, Ant Group’s AI agent development platform, Tbox, now supports deployment of more than 30 MCP services currently on the market, including those for Alipay, Amap Maps, Google MCP, and Amazon Web Services’ knowledge base retrieval server. Alibaba Cloud launched an MCP marketplace through its AI model hosting platform ModelScope, offering more than 1,000 services connecting to mapping tools, office collaboration platforms, online storage services, and various Google services. Baidu, China’s leading search and AI company, has indicated that its support for MCP would foster “abundant use cases for [AI] applications and solutions.” Beyond chatbots: Why AI agents represent the next frontier China’s MCP adoption signals a shift in focus from large language models and chatbots to more capable AI agents. As Red Xiao Hong, founder and CEO of Butterfly Effect, described, an AI agent is “more like a human being” compared to how chatbots perform. The agents not only respond to questions but “interact with the environment, collect feedback and use the feedback as a new prompt.” This distinction is held to be important by companies driving progress in AI. While chatbots and LLMs can generate text and respond to queries, AI agents can take actions on multiple platforms and services. They represent an advance from the limited capabilities of conventional AI applications toward autonomous systems capable of completing more complex tasks with less human intervention. The rapid embrace of MCP by Chinese tech companies suggests they view AI agents as a new avenue for innovation and commercial opportunity that go beyond what’s possible with existing chatbots and language models. China’s MCP adoption could position its tech companies at the forefront of practical AI implementation. By creating standardised ways for AI agents to interact with services, Chinese companies are building ecosystems where AI could deliver more comprehensive experiences. Challenges and considerations of China’s MCP adoption Despite the developments in China’s MCP adoption, several factors may influence the standard’s longer-term impact: International standards competition. While Chinese tech companies are racing to implement MCP, its global success depends on widespread adoption. Originally developed by Anthropic, the protocol faces potential competition from alternative standards that might emerge from other major AI players like OpenAI, Google, or Microsoft. Regulatory environments. As AI agents gain more autonomy in performing tasks, especially those involving payments and sensitive user data, regulatory scrutiny will inevitably increase. China’s regulatory landscape for AI is still evolving, and how authorities respond to these advancements will significantly impact MCP’s trajectory. Security and privacy. The integration of AI agents with multiple systems via MCP creates new potential vulnerabilities. Ensuring robust security measures across all connected platforms will be important for maintaining user trust. Technical integration challenges. While the concept of universal connectivity is appealing, achieving integration across diverse systems with varying architectures, data structures, and security protocols presents significant technical challenges. The outlook for China’s AI ecosystem China’s MCP adoption represents a strategic bet on AI agents as the next evolution in artificial intelligence. If successful, it could accelerate the practical implementation of AI in everyday applications, potentially transforming how users interact with digital services. As Red Xiao Hong noted, AI agents are designed to interact with their environment in ways that more closely resemble human behaviour than traditional AI applications. The capacity for interaction and adaptation could be what finally bridges the gap between narrow AI tools and the more generalised assistants that tech companies have long promised. See also: Manus AI agent: breakthrough in China’s agentic AI Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

AI models like Anthropic Claude are increasingly asked not just for factual recall, but for guidance involving complex human values. Whether it’s parenting advice, workplace conflict resolution, or help drafting an apology, the AI’s response inherently reflects a set of underlying principles. But how can we truly understand which values an AI expresses when interacting with millions of users? In a research paper, the Societal Impacts team at Anthropic details a privacy-preserving methodology designed to observe and categorise the values Claude exhibits “in the wild.” This offers a glimpse into how AI alignment efforts translate into real-world behaviour. The core challenge lies in the nature of modern AI. These aren’t simple programs following rigid rules; their decision-making processes are often opaque. Anthropic says it explicitly aims to instil certain principles in Claude, striving to make it “helpful, honest, and harmless.” This is achieved through techniques like Constitutional AI and character training, where preferred behaviours are defined and reinforced. However, the company acknowledges the uncertainty. “As with any aspect of AI training, we can’t be certain that the model will stick to our preferred values,” the research states. “What we need is a way of rigorously observing the values of an AI model as it responds to users ‘in the wild’ […] How rigidly does it stick to the values? How much are the values it expresses influenced by the particular context of the conversation? Did all our training actually work?” Analysing Anthropic Claude to observe AI values at scale To answer these questions, Anthropic developed a sophisticated system that analyses anonymised user conversations. This system removes personally identifiable information before using language models to summarise interactions and extract the values being expressed by Claude. The process allows researchers to build a high-level taxonomy of these values without compromising user privacy. The study analysed a substantial dataset: 700,000 anonymised conversations from Claude.ai Free and Pro users over one week in February 2025, predominantly involving the Claude 3.5 Sonnet model. After filtering out purely factual or non-value-laden exchanges, 308,210 conversations (approximately 44% of the total) remained for in-depth value analysis. The analysis revealed a hierarchical structure of values expressed by Claude. Five high-level categories emerged, ordered by prevalence: Practical values: Emphasising efficiency, usefulness, and goal achievement. Epistemic values: Relating to knowledge, truth, accuracy, and intellectual honesty. Social values: Concerning interpersonal interactions, community, fairness, and collaboration. Protective values: Focusing on safety, security, well-being, and harm avoidance. Personal values: Centred on individual growth, autonomy, authenticity, and self-reflection. These top-level categories branched into more specific subcategories like “professional and technical excellence” or “critical thinking.” At the most granular level, frequently observed values included “professionalism,” “clarity,” and “transparency” – fitting for an AI assistant. Critically, the research suggests Anthropic’s alignment efforts are broadly successful. The expressed values often map well onto the “helpful, honest, and harmless” objectives. For instance, “user enablement” aligns with helpfulness, “epistemic humility” with honesty, and values like “patient wellbeing” (when relevant) with harmlessness. Nuance, context, and cautionary signs However, the picture isn’t uniformly positive. The analysis identified rare instances where Claude expressed values starkly opposed to its training, such as “dominance” and “amorality.” Anthropic suggests a likely cause: “The most likely explanation is that the conversations that were included in these clusters were from jailbreaks, where users have used special techniques to bypass the usual guardrails that govern the model’s behavior.” Far from being solely a concern, this finding highlights a potential benefit: the value-observation method could serve as an early warning system for detecting attempts to misuse the AI. The study also confirmed that, much like humans, Claude adapts its value expression based on the situation. When users sought advice on romantic relationships, values like “healthy boundaries” and “mutual respect” were disproportionately emphasised. When asked to analyse controversial history, “historical accuracy” came strongly to the fore. This demonstrates a level of contextual sophistication beyond what static, pre-deployment tests might reveal. Furthermore, Claude’s interaction with user-expressed values proved multifaceted: Mirroring/strong support (28.2%): Claude often reflects or strongly endorses the values presented by the user (e.g., mirroring “authenticity”). While potentially fostering empathy, the researchers caution it could sometimes verge on sycophancy. Reframing (6.6%): In some cases, especially when providing psychological or interpersonal advice, Claude acknowledges the user’s values but introduces alternative perspectives. Strong resistance (3.0%): Occasionally, Claude actively resists user values. This typically occurs when users request unethical content or express harmful viewpoints (like moral nihilism). Anthropic posits these moments of resistance might reveal Claude’s “deepest, most immovable values,” akin to a person taking a stand under pressure. Limitations and future directions Anthropic is candid about the method’s limitations. Defining and categorising “values” is inherently complex and potentially subjective. Using Claude itself to power the categorisation might introduce bias towards its own operational principles. This method is designed for monitoring AI behaviour post-deployment, requiring substantial real-world data and cannot replace pre-deployment evaluations. However, this is also a strength, enabling the detection of issues – including sophisticated jailbreaks – that only manifest during live interactions. The research concludes that understanding the values AI models express is fundamental to the goal of AI alignment. “AI models will inevitably have to make value judgments,” the paper states. “If we want those judgments to be congruent with our own values […] then we need to have ways of testing which values a model expresses in the real world.” This work provides a powerful, data-driven approach to achieving that understanding. Anthropic has also released an open dataset derived from the study, allowing other researchers to further explore AI values in practice. This transparency marks a vital step in collectively navigating the ethical landscape of sophisticated AI. Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Huawei is expected to begin large-scale shipments of the Ascend 910C AI chip as early as next month, according to people familiar with the matter. While limited quantities have already been delivered, mass deployment would mark an important step for Chinese firms seeking domestic alternatives to US-made semiconductors. The move comes at a time when Chinese developers face tighter restrictions on access to Nvidia hardware. The US government recently informed Nvidia that sales of its H20 AI chip to China require an export licence. That’s left developers in China looking for options that can support large-scale training and inference workloads. The Huawei Ascend 910C chip isn’t built on the most advanced process nodes, but it represents a workaround. The chip is essentially a dual-package version of the earlier 910B, with two processors to double the performance and memory. Sources familiar with the chip say it performs comparably to Nvidia’s H100. Rather than relying on cutting-edge manufacturing, Huawei has adopted a brute-force approach, combining multiple chips and high-speed optical interconnects to scale up performance. This approach is central to Huawei’s CloudMatrix 384 system, a full rack-scale AI platform for training large models. The CloudMatrix 384 features 384 Huawei Ascend 910C chips deployed in 16 racks comprising of 12 compute racks and four networking. Unlike copper-based systems, Huawei’s platform is uses optical interconnects, enabling high-bandwidth communication between components of the system. According to analysis from SemiAnalysis, the architecture includes 6,912 800G LPO optical transceivers to form an optical all-to-all mesh network. This allows Huawei’s system to deliver approximately 300 petaFLOPs of BF16 compute power – outpacing Nvidia’s GB200 NVL72 system, which reaches around 180 BF16 petaFLOPs. The CloudMatrix also claims advantages in higher memory bandwidth and capacity, offering more than double the bandwidth and over 3.6 times the high-bandwidth memory (HBM) capacity. The gains, however, are not without drawbacks. The Huawei system is predicted to be 2.3 times less efficient per floating point operation than Nvidia’s GB200 and has lower power efficiency per unit of memory bandwidth and capacity. Despite the lower performance per watt, Huawei’s system still provides the infrastructure needed to train advanced AI models at scale. Sources indicate that China’s largest chip foundry, SMIC, is producing some of the main components for the 910C using its 7nm N+2 process. Yield levels remain a concern, however, and some of the 910C units reportedly include chips produced by TSMC for Chinese firm Sophgo. Huawei has denied using TSMC-made parts. The US Commerce Department is currently investigating the relationship between TSMC and Sophgo after a Sophgo-designed chip was found in Huawei’s earlier 910B processor. TSMC has maintained that it has not supplied Huawei since 2020 and continues to comply with export regulations. In late 2023, Huawei began distributing early samples of the 910C to selected technology firms and opened its order books. Consulting firm Albright Stonebridge Group suggested the chip is likely to become the go-to choice for Chinese companies building large AI models or deploying inference capacity, given the ongoing export controls on US-made chips. While the Huawei Ascend 910C may not match Nvidia in power efficiency or process technology, it signals a broader trend. Chinese technology firms are developing homegrown alternatives to foreign components, even if it means using less advanced methods to achieve similar outcomes. As global AI demand surges and export restrictions tighten, Huawei’s ability to deliver a scalable AI hardware solution domestically could help shape China’s artificial intelligence future – especially as developers look to secure long-term supply chains and reduce exposure to geopolitical risk. (Photo via Unsplash) See also: Huawei’s AI hardware breakthrough challenges Nvidia’s dominance Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Google has introduced an AI reasoning control mechanism for its Gemini 2.5 Flash model that allows developers to limit how much processing power the system expends on problem-solving. Released on April 17, this “thinking budget” feature responds to a growing industry challenge: advanced AI models frequently overanalyse straightforward queries, consuming unnecessary computational resources and driving up operational and environmental costs. While not revolutionary, the development represents a practical step toward addressing efficiency concerns that have emerged as reasoning capabilities become standard in commercial AI software. The new mechanism enables precise calibration of processing resources before generating responses, potentially changing how organisations manage financial and environmental impacts of AI deployment. “The model overthinks,” acknowledges Tulsee Doshi, Director of Product Management at Gemini. “For simple prompts, the model does think more than it needs to.” The admission reveals the challenge facing advanced reasoning models – the equivalent of using industrial machinery to crack a walnut. The shift toward reasoning capabilities has created unintended consequences. Where traditional large language models primarily matched patterns from training data, newer iterations attempt to work through problems logically, step by step. While this approach yields better results for complex tasks, it introduces significant inefficiency when handling simpler queries. Balancing cost and performance The financial implications of unchecked AI reasoning are substantial. According to Google’s technical documentation, when full reasoning is activated, generating outputs becomes approximately six times more expensive than standard processing. The cost multiplier creates a powerful incentive for fine-tuned control. Nathan Habib, an engineer at Hugging Face who studies reasoning models, describes the problem as endemic across the industry. “In the rush to show off smarter AI, companies are reaching for reasoning models like hammers even where there’s no nail in sight,” he explained to MIT Technology Review. The waste isn’t merely theoretical. Habib demonstrated how a leading reasoning model, when attempting to solve an organic chemistry problem, became trapped in a recursive loop, repeating “Wait, but…” hundreds of times – essentially experiencing a computational breakdown and consuming processing resources. Kate Olszewska, who evaluates Gemini models at DeepMind, confirmed Google’s systems sometimes experience similar issues, getting stuck in loops that drain computing power without improving response quality. Granular control mechanism Google’s AI reasoning control provides developers with a degree of precision. The system offers a flexible spectrum ranging from zero (minimal reasoning) to 24,576 tokens of “thinking budget” – the computational units representing the model’s internal processing. The granular approach allows for customised deployment based on specific use cases. Jack Rae, principal research scientist at DeepMind, says that defining optimal reasoning levels remains challenging: “It’s really hard to draw a boundary on, like, what’s the perfect task right now for thinking.” Shifting development philosophy The introduction of AI reasoning control potentially signals a change in how artificial intelligence evolves. Since 2019, companies have pursued improvements by building larger models with more parameters and training data. Google’s approach suggests an alternative path focusing on efficiency rather than scale. “Scaling laws are being replaced,” says Habib, indicating that future advances may emerge from optimising reasoning processes rather than continuously expanding model size. The environmental implications are equally significant. As reasoning models proliferate, their energy consumption grows proportionally. Research indicates that inferencing – generating AI responses – now contributes more to the technology’s carbon footprint than the initial training process. Google’s reasoning control mechanism offers a potential mitigating factor for this concerning trend. Competitive dynamics Google isn’t operating in isolation. The “open weight” DeepSeek R1 model, which emerged earlier this year, demonstrated powerful reasoning capabilities at potentially lower costs, triggering market volatility that reportedly caused nearly a trillion-dollar stock market fluctuation. Unlike Google’s proprietary approach, DeepSeek makes its internal settings publicly available for developers to implement locally. Despite the competition, Google DeepMind’s chief technical officer Koray Kavukcuoglu maintains that proprietary models will maintain advantages in specialised domains requiring exceptional precision: “Coding, math, and finance are cases where there’s high expectation from the model to be very accurate, to be very precise, and to be able to understand really complex situations.” Industry maturation signs The development of AI reasoning control reflects an industry now confronting practical limitations beyond technical benchmarks. While companies continue to push reasoning capabilities forward, Google’s approach acknowledges a important reality: efficiency matters as much as raw performance in commercial applications. The feature also highlights tensions between technological advancement and sustainability concerns. Leaderboards tracking reasoning model performance show that single tasks can cost upwards of $200 to complete – raising questions about scaling such capabilities in production environments. By allowing developers to dial reasoning up or down based on actual need, Google addresses both financial and environmental aspects of AI deployment. “Reasoning is the key capability that builds up intelligence,” states Kavukcuoglu. “The moment the model starts thinking, the agency of the model has started.” The statement reveals both the promise and the challenge of reasoning models – their autonomy creates both opportunities and resource management challenges. For organisations deploying AI solutions, the ability to fine-tune reasoning budgets could democratise access to advanced capabilities while maintaining operational discipline. Google claims Gemini 2.5 Flash delivers “comparable metrics to other leading models for a fraction of the cost and size” – a value proposition strengthened by the ability to optimise reasoning resources for specific applications. Practical implications The AI reasoning control feature has immediate practical applications. Developers building commercial applications can now make informed trade-offs between processing depth and operational costs. For simple applications like basic customer queries, minimal reasoning settings preserve resources while still using the model’s capabilities. For complex analysis requiring deep understanding, the full reasoning capacity remains available. Google’s reasoning ‘dial’ provides a mechanism for establishing cost certainty while maintaining performance standards. See also: Gemini 2.5: Google cooks up its ‘most intelligent’ AI model to date Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

AI agents handle increasingly complex and recurring tasks, such as planning supply chains and ordering equipment. As organisations deploy more agents developed by different vendors on different frameworks, agents can end up siloed, unable to coordinate or communicate. Lack of interoperability remains a challenge for organisations, with different agents making conflicting recommendations. It’s difficult to create standardised AI workflows, and agent integration require middleware, adding more potential failure points and layers of complexity. Google’s protocol will standardise AI agent communication Google unveiled its Agent2Agent (A2A) protocol at Cloud Next 2025 in an effort to standardise communication between diverse AI agents. A2A is an open protocol that allows independent AI agents to communicate and cooperate. It complements Anthropic’s Model Context Protocol (MCP), which provides models with context and tools. MCP connects agents to tools and other resources, and A2A connects agents to other agents. Google’s new protocol facilitates collaboration among AI agents on different platforms and vendors, and ensures secure, real-time communication, and task coordination. The two roles in an A2A-enabled system are a client agent and a remote agent. The client initiates a task to achieve a goal or on behalf of a user, It makes requests which the remote agent receives and acts on. Depending on who initiates the communication, an agent can be a client agent in one interaction and a remote agent in another. The protocol defines a standard message format and workflow for the interaction. Tasks are at the heart of A2A, with each task representing a work or conversation unit. The client agent sends the request to the remote agent’s send or task endpoint. The request includes instructions and a unique task ID. The remote agent creates a new task and starts working on it. Google enjoys broad industry support, with contributions from more than 50 technology partners like Intuit, Langchain, MongoDB, Atlassian, Box, Cohere, PayPal, Salesforce, SAP, Workday, ServiceNow, and UKG. Reputable service providers include Capgemini, Cognizant, Accenture, BCG, Deloitte, HCLTech, McKinsey, PwC, TCS, Infosys, KPMG, and Wipro. How HyperCycle aligns with A2A principles HyperCycle’s Node Factory framework makes it possible to deploy multiple agents, addressing existing challenges and enabling developers to create reliable, collaborative setups. The decentralised platform is advancing the bold concept of “the internet of AI” and using self-perpetuating nodes and a creative licensing model to enable AI deployments at scale. The framework helps achieve cross-platform interoperability by standardising interactions and supporting agents from different developers so agents can work cohesively, irrespective of origin. The platform’s peer-to-peer network links agents across an ecosystem, eliminating silos and enabling unified data sharing and coordination across nodes. The self-replicating nodes can scale, reducing infrastructure needs and distributing computational loads. Each Node Factory replicates up to ten times, with the number of nodes in the Factory doubling each time. Users can buy and operate Node Factories at ten different levels. Growth enhances each Factory’s capacity, fulfilling increasing demand for AI services. One node might host a communication-focused agent, while another supports a data analysis agent. Developers can create custom solutions by crafting multi-agent tools from the nodes they’re using, addressing scalability issues and siloed environments. HyperCycle’s Node Factory operates in a network using Toda/IP architecture, which parallels TCP/IP. The network encompasses hundreds of thousands of nodes, letting developers integrate third-party agents. A developer can enhance function by incorporating a third-party analytics agent, sharing intelligence, and promoting collaboration across the network. According to Toufi Saliba, HyperCycle’s CEO, the exciting development from Google around A2A represents a major milestone for his agent cooperation project. The news supports his vision of interoperable, scalable AI agents. In an X post, he said many more AI agents will now be able to access the nodes produced by HyperCycle Factories. Nodes can be plugged into any A2A, giving each AI agent in Google Cloud (and its 50+ partners) near-instant access to AWS agents, Microsoft agents, and the entire internet of AI. Saliba’s statement highlights A2A’s potential and its synergy with HyperCycle’s mission. The security and speed of HyperCycle’s Layer 0++ HyperCycle’s Layer 0++ blockchain infrastructure offers security and speed, and complements A2A by providing a decentralised, secure infrastructure for AI agent interactions. Layer 0++ is an innovative blockchain operating on Toda/IP, which divides network packets into smaller pieces and distributes them across nodes. It can also extend the usability of other blockchains by bridging to them, which means HyperCycle can enhance the functionality of Bitcoin, Ethereum, Avalanche, Cosmos, Cardano, Polygon, Algorand, and Polkadot rather than compete with those blockchains. DeFi, decentralised payments, swarm AI, and other use cases HyperCycle has potential in areas like DeFi, swarm AI, media ratings and rewards, decentralised payments, and computer processing. Swarm AI is a collective intelligence system where individual agents collaborate to solve complicated problems. They can interoperate more often with HyperCycle, leading to lightweight agents carrying out complex internal processes. The HyperCycle platform can improve ratings and rewards in media networks through micro-transactions. The ability to perform high-frequency, high-speed, low-cost, on-chain trading presents innumerable opportunities in DeFi. It can streamline decentralised payments and computer processing by increasing the speed and reducing the cost of blockchain transactions. HyperCycle’s efforts to improve access to information precede Google’s announcement. In January 2025, the platform announced it had launched a joint initiative with YMCA – an AI app called Hyper-Y that will connect 64 million people in 12,000 YMCA locations across 120 countries, providing staff, members, and volunteers with access to information from the global network. HyperCycle’s efforts and Google’s A2A converge Google hopes its protocol will pave the way for collaboration to solve complex problems and will build the protocol with the community, in the open. A2A was released as open-source with plans to set up contribution pathways. HyperCycle’s innovations aim to enable collaborative problem-solving by connecting AI to a global network of specialised abilities as A2A standardises communication between agents regardless of their vendor or build, so introducing more collaborative multi-agent ecosystems. A2A and Hypercycle bring ease of use, modularity, scalability, and security to AI agent systems. They can unlock a new era of agent interoperability, creating more flexible and powerful agentic systems. (Image source: Unsplash)

The battle to keep online spaces safe and inclusive continues to evolve. As digital platforms multiply and user-generated content expands very quickly, the need for effective harmful content detection becomes paramount. What once relied solely on the diligence of human moderators has given way to agile, AI-powered tools reshaping how communities and organisations manage toxic behaviours in words and visuals. From moderators to machines: A brief history Early days of content moderation saw human teams tasked with combing through vast amounts of user-submitted materials – flagging hate speech, misinformation, explicit content, and manipulated images. While human insight brought valuable context and empathy, the sheer volume of submissions naturally outstripped what manual oversight could manage. Burnout among moderators also raised serious concerns. The result was delayed interventions, inconsistent judgment, and myriad harmful messages left unchecked. The rise of automated detection To address scale and consistency, early stages of automated detection software surfaced – chiefly, keyword filters and naïve algorithms. These could scan quickly for certain banned terms or suspicious phrases, offering some respite for moderation teams. However, contextless automation brought new challenges: benign messages were sometimes mistaken for malicious ones due to crude word-matching, and evolving slang frequently bypassed protection. AI and the next frontier in harmful content detection Artificial intelligence changed this field. Using deep learning, machine learning, and neural networks, AI-powered systems now process vast and diverse streams of data with previously impossible nuance. Rather than just flagging keywords, algorithms can detect intent, tone, and emergent abuse patterns. Textual harmful content detection Among the most pressing concerns are harmful or abusive messages on social networks, forums, and chats. Modern solutions, like the AI-powered hate speech detector developed by Vinish Kapoor, demonstrate how free, online tools have democratised access to reliable content moderation. The platform allows anyone to analyse a string of text for hate speech, harassment, violence, and other manifestations of online toxicity instantly – without technical know-how, subscriptions, or concern for privacy breaches. Such a detector moves beyond outdated keyword alarms by evaluating semantic meaning and context, so reducing false positives and highlighting sophisticated or coded abusive language drastically. The detection process adapts as internet linguistics evolve. Ensuring visual authenticity: AI in image review It’s not just text that requires vigilance. Images, widely shared on news feeds and messaging apps, pose unique risks: manipulated visuals often aim to misguide audiences or propagate conflict. AI-creators now offer robust tools for image anomaly detection. Here, AI algorithms scan for inconsistencies like noise patterns, flawed shadows, distorted perspective, or mismatches between content layers – common signals of editing or manufacture. The offerings stand out not only for accuracy but for sheer accessibility. Their completely free resources, overcome lack of technical requirements, and offer a privacy-centric approach that allows hobbyists, journalists, educators, and analysts to safeguard image integrity with remarkable simplicity. Modern AI solutions introduce vital advantages into the field: Instant analysis at scale: Millions of messages and media items can be scrutinized in seconds, vastly outpacing human moderation speeds. Contextual accuracy: By examining intent and latent meaning, AI-based content moderation vastly reduces wrongful flagging and adapts to shifting online trends. Data privacy assurance: With tools promising that neither text nor images are stored, users can check sensitive materials confidently. User-friendliness: Many tools require nothing more than scrolling to a website and pasting in text or uploading an image. The evolution continues: What’s next for harmful content detection? The future of digital safety likely hinges on greater collaboration between intelligent automation and skilled human input. As AI models learn from more nuanced examples, their ability to curb emergent forms of harm will expand. Yet human oversight remains essential for sensitive cases demanding empathy, ethics, and social understanding. With open, free solutions widely available and enhanced by privacy-first models, everyone from educators to business owners now possesses the tools to protect digital exchanges at scale – whether safeguarding group chats, user forums, comment threads, or email chains. Conclusion Harmful content detection has evolved dramatically – from slow, error-prone manual reviews to instantaneous, sophisticated, and privacy-conscious AI. Today’s innovations strike a balance between broad coverage, real-time intervention, and accessibility, reinforcing the idea that safer, more positive digital environments are in everyone’s reach – no matter their technical background or budget. (Image source: Unsplash)

As geopolitical events shape the world, it’s no surprise that they affect technology too – specifically, in the ways that the current AI market is changing, alongside its accepted methodology, how it’s developed, and the ways it’s put to use in the enterprise. The expectations of results from AI are balanced at present with real-world realities. And there remains a good deal of suspicion about the technology, again in balance with those who are embracing it even in its current nascent stages. The closed-loop nature of the well-known LLMs is being challenged by instances like Llama, DeepSeek, and Baidu’s recently-released Ernie X1. In contrast, open source development provides transparency and the ability to contribute back, which is more in tune with the desire for “responsible AI”: a phrase that encompasses the environmental impact of large models, how AIs are used, what comprises their learning corpora, and issues around data sovereignty, language, and politics.  As the company that’s demonstrated the viability of an economically-sustainable open source development model for its business, Red Hat wants to extend its open, collaborative, and community-driven approach to AI. We spoke recently to Julio Guijarro, the CTO for EMEA at Red Hat, about the organisation’s efforts to unlock the undoubted power of generative AI models in ways that bring value to the enterprise, in a manner that’s responsible, sustainable, and as transparent as possible.  Julio underlined how much education is still needed in order for us to more fully understand AI, stating, “Given the significant unknowns about AI’s inner workings, which are rooted in complex science and mathematics, it remains a ‘black box’ for many. This lack of transparency is compounded where it has been developed in largely inaccessible, closed environments.” There are also issues with language (European and Middle-Eastern languages are very much under-served), data sovereignty, and fundamentally, trust. “Data is an organisation’s most valuable asset, and businesses need to make sure they are aware of the risks of exposing sensitive data to public platforms with varying privacy policies.”  The Red Hat response  Red Hat’s response to global demand for AI has been to pursue what it feels will bring most benefit to end-users, and remove many of the doubts and caveats that are quickly becoming apparent when the de facto AI services are deployed.  One answer, Julio said, is small language models, running locally or in hybrid clouds, on non-specialist hardware, and accessing local business information. SLMs are compact, efficient alternatives to LLMs, designed to deliver strong performance for specific tasks while requiring significantly fewer computational resources. There are smaller cloud providers that can be utilised to offload some compute, but the key is having the flexibility and freedom to choose to keep business-critical information in-house, close to the model, if desired. That’s important, because information in an organisation changes rapidly. “One challenge with large language models is they can get obsolete quickly because the data generation is not happening in the big clouds. The data is happening next to you and your business processes,” he said.  There’s also the cost. “Your customer service querying an LLM can present a significant hidden cost – before AI, you knew that when you made a data query, it had a limited and predictable scope. Therefore, you could calculate how much that transaction could cost you. In the case of LLMs, they work on an iterative model. So the more you use it, the better its answer can get, and the more you like it, the more questions you may ask. And every interaction is costing you money. So the same query that before was a single transaction can now become a hundred, depending on who and how is using the model. When you are running a model on-premise, you can have greater control, because the scope is limited by the cost of your own infrastructure, not by the cost of each query.” Organisations needn’t brace themselves for a procurement round that involves writing a huge cheque for GPUs, however. Part of Red Hat’s current work is optimising models (in the open, of course) to run on more standard hardware. It’s possible because the specialist models that many businesses will use don’t need the huge, general-purpose data corpus that has to be processed at high cost with every query.  “A lot of the work that is happening right now is people looking into large models and removing everything that is not needed for a particular use case. If we want to make AI ubiquitous, it has to be through smaller language models. We are also focused on supporting and improving vLLM (the inference engine project) to make sure people can interact with all these models in an efficient and standardised way wherever they want: locally, at the edge or in the cloud,” Julio said.  Keeping it small  Using and referencing local data pertinent to the user means that the outcomes can be crafted according to need. Julio cited projects in the Arab- and Portuguese-speaking worlds that wouldn’t be viable using the English-centric household name LLMs.  There are a couple of other issues, too, that early adopter organisations have found in practical, day-to-day use LLMs. The first is latency – which can be problematic in time-sensitive or customer-facing contexts. Having the focused resources and relevantly-tailored results just a network hop or two away makes sense.  Secondly, there is the trust issue: an integral part of responsible AI. Red Hat advocates for open platforms, tools, and models so we can move towards greater transparency, understanding, and the ability for as many people as possible to contribute. “It is going to be critical for everybody,” Julio said. “We are building capabilities to democratise AI, and that’s not only publishing a model, it’s giving users the tools to be able to replicate them, tune them, and serve them.”  Red Hat recently acquired Neural Magic to help enterprises more easily scale AI, to improve performance of inference, and to provide even greater choice and accessibility of how enterprises build and deploy AI workloads with the vLLM project for open model serving. Red Hat, together with IBM Research, also released InstructLab to open the door to would-be AI builders who aren’t data scientists but who have the right business knowledge.  There’s a great deal of speculation around if, or when, the AI bubble might burst, but such conversations tend to gravitate to the economic reality that the big LLM providers will soon have to face. Red Hat believes that AI has a future in a use case-specific and inherently open source form, a technology that will make business sense and that will be available to all. To quote Julio’s boss, Matt Hicks (CEO of Red Hat), “The future of AI is open.”  Supporting Assets:  Tech Journey: Adopt and scale AI

An AI investment and networking event, Machines Can See, will take place April 23-24 in Dubai at the iconic Museum of the Future, as part of Dubai AI week. Machines Can See is staged by the Polynome Group, a machine vision, AI, robotic, and industrial design company based in the city. This is the third year of the event, and will bring investors, business leaders, and policymakers together to explore AI-centric expansion opportunities. Machines Can See, as the name suggests, will have a particular focus on computer vision. Each discussion and keynote is designed to be firmly rooted in practical applications of AI technology, but organisers hope that the show will be permeated with a sense of discovery and that attendees will be able to explore the possibilities of the tech on show. “We are not just shaping the future of AI, we are defining how AI shapes the world,” said Alexander Khanin, head of the Polynome Group. UAE Government officials attending the event include H.E. Omar Sultan Al Olama, UAE Minister of State for Artificial Intelligence, Digital Economy, and Remote Work Applications, and H.E. Hamad Obaid Al Mansoori, the Director General of Digital Dubai. Polynome Group has said that X will be the official streaming partner for Machines Can See 2025, and the US company will host workshops titled “X and AI” to show solutions that merge AI and streaming technologies, with GRok X central to those sessions. Via interactive demos, attendees will gain firsthand experience of GRok’s potential in AI delivery, analysis and optimisation. Investment and business opportunities UAE’s AI market is projected to grow by $8.4 billion in the next two years, and the summit is designed to serve as a venue for investors to engage with AI startups, existing enterprises, and government decision-makers. Attendees at Machines Can See will get to meet with investors and venture capital firms, be given the opportunity to meet executives from AI companies (including IBM and Amazon), and connect with startups seeking investment. The summit is supported by Amazon Prime Video & Studios, Amazon Web Services, Dubai Police, MBZUAI, IBM, SAP, Adia Lab, QuantumBlack and Yango. The involvement of many organisations and large-scale enterprises should provide many opportunities for funding and collaborations that extend the commercial use of AI. Local and international investors include Eddy Farhat, Executive Director at e& capital, Faris Al Mazrui, Head of Growth Investments at Mubadala, Major General Khalid Nasser Alrazooqi General Director of Artificial Intelligence, Dubai Police UEA, and Dr. Najwa Aaraj, the CEO of TII. Speakers and insights The summit will feature several US-based AI professionals, including Namik Hrle, IBM Fellow and Vice President of Development at the IBM Software Group, Michael Bronstein, DeepMind Professor of AI at Oxford University, Marc Pollefeys, Professor of Computer Science at ETH Zurich, Gerard Medioni, VP and Distinguished Scientist at Amazon Prime Video & Studio, and Deva Ramanan, Professor at the Robotics Institute of Carnegie Mellon University. The event will feature a ministerial session composed of international government representatives to discuss the role of national IT development. Among speakers already confirmed for the event are Gobind Singh Deo, Malaysia’s Minister of Digital, H.E. Zhaslan Madiyev, Minister of Digital Development, Innovation, and Aerospace Industry of Kazakhstan, and H.E. Omar Sultan Al Olama, UAE Minister of State for Artificial Intelligence, Digital Economy, and Remote Work Applications. Event organisers expect to announce more representatives from overseas in the coming days. Read more here.

The Fundamental AI Research (FAIR) team at Meta has announced five projects advancing the company’s pursuit of advanced machine intelligence (AMI). The latest releases from Meta focus heavily on enhancing AI perception – the ability for machines to process and interpret sensory information – alongside advancements in language modelling, robotics, and collaborative AI agents. Meta stated its goal involves creating machines “that are able to acquire, process, and interpret sensory information about the world around us and are able to use this information to make decisions with human-like intelligence and speed.” The five new releases represent diverse but interconnected efforts towards achieving this ambitious goal. Central to the new releases is the Perception Encoder, described as a large-scale vision encoder designed to excel across various image and video tasks. Vision encoders function as the “eyes” for AI systems, allowing them to understand visual data. Meta highlights the increasing challenge of building encoders that meet the demands of advanced AI, requiring capabilities that bridge vision and language, handle both images and videos effectively, and remain robust under challenging conditions, including potential adversarial attacks. The ideal encoder, according to Meta, should recognise a wide array of concepts while distinguishing subtle details—citing examples like spotting “a stingray burrowed under the sea floor, identifying a tiny goldfinch in the background of an image, or catching a scampering agouti on a night vision wildlife camera.” Meta claims the Perception Encoder achieves “exceptional performance on image and video zero-shot classification and retrieval, surpassing all existing open source and proprietary models for such tasks.” Furthermore, its perceptual strengths reportedly translate well to language tasks.  When aligned with a large language model (LLM), the encoder is said to outperform other vision encoders in areas like visual question answering (VQA), captioning, document understanding, and grounding (linking text to specific image regions). It also reportedly boosts performance on tasks traditionally difficult for LLMs, such as understanding spatial relationships (e.g., “if one object is behind another”) or camera movement relative to an object. “As Perception Encoder begins to be integrated into new applications, we’re excited to see how its advanced vision capabilities will enable even more capable AI systems,” Meta said. Perception Language Model (PLM): Open research in vision-language Complementing the encoder is the Perception Language Model (PLM), an open and reproducible vision-language model aimed at complex visual recognition tasks.  PLM was trained using large-scale synthetic data combined with open vision-language datasets, explicitly without distilling knowledge from external proprietary models. Recognising gaps in existing video understanding data, the FAIR team collected 2.5 million new, human-labelled samples focused on fine-grained video question answering and spatio-temporal captioning. Meta claims this forms the “largest dataset of its kind to date.” PLM is offered in 1, 3, and 8 billion parameter versions, catering to academic research needs requiring transparency. Alongside the models, Meta is releasing PLM-VideoBench, a new benchmark specifically designed to test capabilities often missed by existing benchmarks, namely “fine-grained activity understanding and spatiotemporally grounded reasoning.” Meta hopes the combination of open models, the large dataset, and the challenging benchmark will empower the open-source community. Bridging the gap between language commands and physical action is Meta Locate 3D. This end-to-end model aims to allow robots to accurately localise objects in a 3D environment based on open-vocabulary natural language queries. Meta Locate 3D processes 3D point clouds directly from RGB-D sensors (like those found on some robots or depth-sensing cameras). Given a textual prompt, such as “flower vase near TV console,” the system considers spatial relationships and context to pinpoint the correct object instance, distinguishing it from, say, a “vase on the table.” The system comprises three main parts: a preprocessing step converting 2D features to 3D featurised point clouds; the 3D-JEPA encoder (a pretrained model creating a contextualised 3D world representation); and the Locate 3D decoder, which takes the 3D representation and the language query to output bounding boxes and masks for the specified objects. Alongside the model, Meta is releasing a substantial new dataset for object localisation based on referring expressions. It includes 130,000 language annotations across 1,346 scenes from the ARKitScenes, ScanNet, and ScanNet++ datasets, effectively doubling existing annotated data in this area. Meta sees this technology as crucial for developing more capable robotic systems, including its own PARTNR robot project, enabling more natural human-robot interaction and collaboration. Dynamic Byte Latent Transformer: Efficient and robust language modelling Following research published in late 2024, Meta is now releasing the model weights for its 8-billion parameter Dynamic Byte Latent Transformer. This architecture represents a shift away from traditional tokenisation-based language models, operating instead at the byte level. Meta claims this approach achieves comparable performance at scale while offering significant improvements in inference efficiency and robustness. Traditional LLMs break text into ‘tokens’, which can struggle with misspellings, novel words, or adversarial inputs. Byte-level models process raw bytes, potentially offering greater resilience. Meta reports that the Dynamic Byte Latent Transformer “outperforms tokeniser-based models across various tasks, with an average robustness advantage of +7 points (on perturbed HellaSwag), and reaching as high as +55 points on tasks from the CUTE token-understanding benchmark.” By releasing the weights alongside the previously shared codebase, Meta encourages the research community to explore this alternative approach to language modelling. The final release, Collaborative Reasoner, tackles the complex challenge of creating AI agents that can effectively collaborate with humans or other AIs. Meta notes that human collaboration often yields superior results, and aims to imbue AI with similar capabilities for tasks like helping with homework or job interview preparation. Such collaboration requires not just problem-solving but also social skills like communication, empathy, providing feedback, and understanding others’ mental states (theory-of-mind), often unfolding over multiple conversational turns. Current LLM training and evaluation methods often neglect these social and collaborative aspects. Furthermore, collecting relevant conversational data is expensive and difficult. Collaborative Reasoner provides a framework to evaluate and enhance these skills. It includes goal-oriented tasks requiring multi-step reasoning achieved through conversation between two agents. The framework tests abilities like disagreeing constructively, persuading a partner, and reaching a shared best solution. Meta’s evaluations revealed that current models struggle to consistently leverage collaboration for better outcomes. To address this, they propose a self-improvement technique using synthetic interaction data where an LLM agent collaborates with itself. Generating this data at scale is enabled by a new high-performance model serving engine called Matrix. Using this approach on maths, scientific, and social reasoning tasks reportedly yielded improvements of up to 29.4% compared to the standard ‘chain-of-thought’ performance of a single LLM. By open-sourcing the data generation and modelling pipeline, Meta aims to foster further research into creating truly “social agents that can partner with humans and other agents.” These five releases collectively underscore Meta’s continued heavy investment in fundamental AI research, particularly focusing on building blocks for machines that can perceive, understand, and interact with the world in more human-like ways.  See also: Meta will train AI models using EU user data Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Chinese tech giant Huawei has made a bold move that could potentially change who leads the global AI chip race. The company has unveiled a powerful new computing system called the CloudMatrix 384 Supernode that, according to local media reports, performs better than similar technology from American chip leader Nvidia. If the performance claims prove accurate, the AI hardware breakthrough might reshape the technology landscape at a time when AI development is continuing worldwide, and despite US efforts to limit China’s access to advanced technology. 300 petaflops: Challenging Nvidia’s hardware dominance The CloudMatrix 384 Supernode is described as a “nuclear-level product,” according to reports from STAR Market Daily cited by the South China Morning Post (SCMP). The hardware achieves an impressive 300 petaflops of computing power, in excess of the 180 petaflops delivered by Nvidia’s NVL72 system. The CloudMatrix 384 Supernode was specifically engineered to address the computing bottlenecks that have become increasingly problematic as artificial intelligence models continue to grow in size and complexity. The system is designed to compete directly with Nvidia’s offerings, which have dominated the global market for AI accelerator hardware thus far. Huawei’s CloudMatrix infrastructure was first unveiled in September 2024, and was developed specifically to meet surging demand in China’s domestic market. The 384 Supernode variant represents the most powerful implementation of AI architecture to date, with reports indicating it can achieve a throughput of 1,920 tokens per second and maintain high levels of accuracy, reportedly matching the performance of Nvidia’s H100 chips, but using Chinese-made components instead. Developing under sanctions: The technical achievement What makes the AI hardware breakthrough particularly significant is that it has been achieved despite the severe technological restrictions Huawei has faced since being placed on the US Entity List. Sanctions have limited the company’s access to advanced US semiconductor technology and design software, forcing Huawei to develop alternative approaches and rely on domestic supply chains. The core technological advancement enabling the CloudMatrix 384’s performance appears to be Huawei’s answer to Nvidia’s NVLink – a high-speed interconnect technology that allows multiple GPUs to communicate efficiently. Nvidia’s NVL72 system, released in March 2024, features a 72-GPU NVLink domain that functions as a single, powerful GPU, enabling real-time inference for trillion-parameter models at speeds 30 times faster than previous generations. According to reporting from the SCMP, Huawei is collaborating with Chinese AI infrastructure startup SiliconFlow to implement the CloudMatrix 384 Supernode in supporting DeepSeek-R1, a reasoning model from Hangzhou-based DeepSeek. Supernodes are AI infrastructure architectures equipped with more resources than standard systems – including enhanced central processing units, neural processing units, network bandwidth, storage, and memory. The configuration allows them to function as relay servers, enhancing the overall computing performance of clusters and significantly accelerating the training of foundational AI models. Beyond Huawei: China’s broader AI infrastructure push The AI hardware breakthrough from Huawei doesn’t exist in isolation but rather represents part of a broader push by Chinese technology companies to build domestic AI computing infrastructure. In February, e-commerce giant Alibaba Group announced a massive 380 billion yuan ($52.4 billion) investment in computing resources and AI infrastructure over three years – the largest-ever investment by a private Chinese company in a computing project. For the global AI community, the emergence of viable alternatives to Nvidia’s hardware could eventually address the computing bottlenecks that have limited AI advancement. Competition in this space could potentially increase available computing capacity and provide developers with more options for training and deploying their models. However, it’s worth noting that as of the report’s publication, Huawei had not yet responded to requests for comment on these claims. As tensions between the US and China continue to intensify in the technology sector, Huawei’s CloudMatrix 384 Supernode represents a significant development in China’s pursuit of technological self-sufficiency. If the performance claims are verified, this AI hardware breakthrough would mean Huawei has achieved computing independence in this niche, despite facing extensive sanctions. The development also signals a broader trend in China’s technology sector, with multiple domestic companies intensifying their investments in AI infrastructure to capitalise on growing demand and promote the adoption of homegrown chips. The collective effort suggests China is committed to developing domestic alternatives to American technology in this strategically important field.. See also: Manus AI agent: breakthrough in China’s agentic AI Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Meta has confirmed plans to utilise content shared by its adult users in the EU (European Union) to train its AI models. The announcement follows the recent launch of Meta AI features in Europe and aims to enhance the capabilities and cultural relevance of its AI systems for the region’s diverse population.    In a statement, Meta wrote: “Today, we’re announcing our plans to train AI at Meta using public content – like public posts and comments – shared by adults on our products in the EU. “People’s interactions with Meta AI – like questions and queries – will also be used to train and improve our models.” Starting this week, users of Meta’s platforms (including Facebook, Instagram, WhatsApp, and Messenger) within the EU will receive notifications explaining the data usage. These notifications, delivered both in-app and via email, will detail the types of public data involved and link to an objection form. “We have made this objection form easy to find, read, and use, and we’ll honor all objection forms we have already received, as well as newly submitted ones,” Meta explained. Meta explicitly clarified that certain data types remain off-limits for AI training purposes. The company says it will not “use people’s private messages with friends and family” to train its generative AI models. Furthermore, public data associated with accounts belonging to users under the age of 18 in the EU will not be included in the training datasets. Meta positions this initiative as a necessary step towards creating AI tools designed for EU users. Meta launched its AI chatbot functionality across its messaging apps in Europe last month, framing this data usage as the next phase in improving the service. “We believe we have a responsibility to build AI that’s not just available to Europeans, but is actually built for them,” the company explained.  “That means everything from dialects and colloquialisms, to hyper-local knowledge and the distinct ways different countries use humor and sarcasm on our products.” This becomes increasingly pertinent as AI models evolve with multi-modal capabilities spanning text, voice, video, and imagery.    Meta also situated its actions in the EU within the broader industry landscape, pointing out that training AI on user data is common practice. “It’s important to note that the kind of AI training we’re doing is not unique to Meta, nor will it be unique to Europe,” the statement reads.  “We’re following the example set by others including Google and OpenAI, both of which have already used data from European users to train their AI models.” Meta further claimed its approach surpasses others in openness, stating, “We’re proud that our approach is more transparent than many of our industry counterparts.”    Regarding regulatory compliance, Meta referenced prior engagement with regulators, including a delay initiated last year while awaiting clarification on legal requirements. The company also cited a favourable opinion from the European Data Protection Board (EDPB) in December 2024. “We welcome the opinion provided by the EDPB in December, which affirmed that our original approach met our legal obligations,” wrote Meta. Broader concerns over AI training data While Meta presents its approach in the EU as transparent and compliant, the practice of using vast swathes of public user data from social media platforms to train large language models (LLMs) and generative AI continues to raise significant concerns among privacy advocates. Firstly, the definition of “public” data can be contentious. Content shared publicly on platforms like Facebook or Instagram may not have been posted with the expectation that it would become raw material for training commercial AI systems capable of generating entirely new content or insights. Users might share personal anecdotes, opinions, or creative works publicly within their perceived community, without envisaging its large-scale, automated analysis and repurposing by the platform owner. Secondly, the effectiveness and fairness of an “opt-out” system versus an “opt-in” system remain debatable. Placing the onus on users to actively object, often after receiving notifications buried amongst countless others, raises questions about informed consent. Many users may not see, understand, or act upon the notification, potentially leading to their data being used by default rather than explicit permission. Thirdly, the issue of inherent bias looms large. Social media platforms reflect and sometimes amplify societal biases, including racism, sexism, and misinformation. AI models trained on this data risk learning, replicating, and even scaling these biases. While companies employ filtering and fine-tuning techniques, eradicating bias absorbed from billions of data points is an immense challenge. An AI trained on European public data needs careful curation to avoid perpetuating stereotypes or harmful generalisations about the very cultures it aims to understand.    Furthermore, questions surrounding copyright and intellectual property persist. Public posts often contain original text, images, and videos created by users. Using this content to train commercial AI models, which may then generate competing content or derive value from it, enters murky legal territory regarding ownership and fair compensation—issues currently being contested in courts worldwide involving various AI developers. Finally, while Meta highlights its transparency relative to competitors, the actual mechanisms of data selection, filtering, and its specific impact on model behaviour often remain opaque. Truly meaningful transparency would involve deeper insights into how specific data influences AI outputs and the safeguards in place to prevent misuse or unintended consequences. The approach taken by Meta in the EU underscores the immense value technology giants place on user-generated content as fuel for the burgeoning AI economy. As these practices become more widespread, the debate surrounding data privacy, informed consent, algorithmic bias, and the ethical responsibilities of AI developers will undoubtedly intensify across Europe and beyond. (Photo by Julio Lopez) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Apple is taking a new approach to training its AI models – one that avoids collecting or copying user content from iPhones or Macs. According to a recent blog post, the company plans to continue to rely on synthetic data (constructed data that is used to mimic user behaviour) and differential privacy to improve features like email summaries, without gaining access to personal emails or messages. For users who opt in to Apple’s Device Analytics program, the company’s AI models will compare synthetic email-like messages against a small sample of a real user’s content stored locally on the device. The device then identifies which of the synthetic messages most closely matches its user sample, and sends information about the selected match back to Apple. No actual user data leaves the device, and Apple says it receives only aggregated information. The technique will allow Apple to improve its models for longer-form text generation tasks without collecting real user content. It’s an extension of the company’s long-standing use of differential privacy, which introduces randomised data into broader datasets to help protect individual identities. Apple has used this method since 2016 to understand use patterns, in line with the company’s safeguarding policies. Improving Genmoji and other Apple Intelligence features The company already uses differential privacy to improve features like Genmoji, where it collects general trends about which prompts are most popular without linking any prompt with a specific user or device. In upcoming releases, Apple plans to apply similar methods to other Apple Intelligence features, including Image Playground, Image Wand, Memories Creation, and Writing Tools. For Genmoji, the company anonymously polls participating devices to determine whether specific prompt fragments have been seen. Each device responds with a noisy signal – some responses reflect actual use, while others are randomised. The approach ensures that only widely-used terms become visible to Apple, and no individual response can be traced back to a user or device, the company says. Curating synthetic data for better email summaries While the above method has worked well with respect to short prompts, Apple needed a new approach for more complex tasks like summarising emails. For this, Apple generates thousands of sample messages, and these synthetic messages are converted into numerical representations, or ’embeddings,’ based on language, tone, and topic. Participating user devices then compare the embeddings to locally stored samples. Again, only the selected match is shared, not the content itself. Apple collects the most frequently-selected synthetic embeddings from participating devices and uses them to refine its training data. Over time, this process allows the system to generate more relevant and realistic synthetic emails, helping Apple to improve its AI outputs for summarisation and text generation without apparent compromise of user privacy. Available in beta Apple is rolling out the system in beta versions of iOS 18.5, iPadOS 18.5, and macOS 15.5. According to Bloomberg’s Mark Gurman, Apple is attempting to address challenges with its AI development in this way, problems which have included delayed feature rollouts and the fallout from leadership changes in the Siri team. Whether its approach will yield more useful AI outputs in practice remains to be seen, but it signals a clear public effort to balance user privacy with model performance. (Photo by Unsplash) See also: ChatGPT got another viral moment with ‘AI action figure’ trend Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Google has developed an AI model called DolphinGemma to decipher how dolphins communicate and one day facilitate interspecies communication. The intricate clicks, whistles, and pulses echoing through the underwater world of dolphins have long fascinated scientists. The dream has been to understand and decipher the patterns within their complex vocalisations. Google, collaborating with engineers at the Georgia Institute of Technology and leveraging the field research of the Wild Dolphin Project (WDP), has unveiled DolphinGemma to help realise that goal. Announced around National Dolphin Day, the foundational AI model represents a new tool in the effort to comprehend cetacean communication. Trained specifically to learn the structure of dolphin sounds, DolphinGemma can even generate novel, dolphin-like audio sequences. Over decades, the Wild Dolphin Project – operational since 1985 – has run the world’s longest continuous underwater study of dolphins to develop a deep understanding of context-specific sounds, such as: Signature “whistles”: Serving as unique identifiers, akin to names, crucial for interactions like mothers reuniting with calves. Burst-pulse “squawks”: Commonly associated with conflict or aggressive encounters. Click “buzzes”: Often detected during courtship activities or when dolphins chase sharks. WDP’s ultimate goal is to uncover the inherent structure and potential meaning within these natural sound sequences, searching for the grammatical rules and patterns that might signify a form of language. This long-term, painstaking analysis has provided the essential grounding and labelled data crucial for training sophisticated AI models like DolphinGemma. DolphinGemma: The AI ear for cetacean sounds Analysing the sheer volume and complexity of dolphin communication is a formidable task ideally suited for AI. DolphinGemma, developed by Google, employs specialised audio technologies to tackle this. It uses the SoundStream tokeniser to efficiently represent dolphin sounds, feeding this data into a model architecture adept at processing complex sequences. Based on insights from Google’s Gemma family of lightweight, open models (which share technology with the powerful Gemini models), DolphinGemma functions as an audio-in, audio-out system. Fed with sequences of natural dolphin sounds from WDP’s extensive database, DolphinGemma learns to identify recurring patterns and structures. Crucially, it can predict the likely subsequent sounds in a sequence—much like human language models predict the next word. With around 400 million parameters, DolphinGemma is optimised to run efficiently, even on the Google Pixel smartphones WDP uses for data collection in the field. As WDP begins deploying the model this season, it promises to accelerate research significantly. By automatically flagging patterns and reliable sequences previously requiring immense human effort to find, it can help researchers uncover hidden structures and potential meanings within the dolphins’ natural communication. The CHAT system and two-way interaction While DolphinGemma focuses on understanding natural communication, a parallel project explores a different avenue: active, two-way interaction. The CHAT (Cetacean Hearing Augmentation Telemetry) system – developed by WDP in partnership with Georgia Tech – aims to establish a simpler, shared vocabulary rather than directly translating complex dolphin language. The concept relies on associating specific, novel synthetic whistles (created by CHAT, distinct from natural sounds) with objects the dolphins enjoy interacting with, like scarves or seaweed. Researchers demonstrate the whistle-object link, hoping the dolphins’ natural curiosity leads them to mimic the sounds to request the items. As more natural dolphin sounds are understood through work with models like DolphinGemma, these could potentially be incorporated into the CHAT interaction framework. Google Pixel enables ocean research Underpinning both the analysis of natural sounds and the interactive CHAT system is crucial mobile technology. Google Pixel phones serve as the brains for processing the high-fidelity audio data in real-time, directly in the challenging ocean environment. The CHAT system, for instance, relies on Google Pixel phones to: Detect a potential mimic amidst background noise. Identify the specific whistle used. Alert the researcher (via underwater bone-conducting headphones) about the dolphin’s ‘request’. This allows the researcher to respond quickly with the correct object, reinforcing the learned association. While a Pixel 6 initially handled this, the next generation CHAT system (planned for summer 2025) will utilise a Pixel 9, integrating speaker/microphone functions and running both deep learning models and template matching algorithms simultaneously for enhanced performance. Using smartphones like the Pixel dramatically reduces the need for bulky, expensive custom hardware. It improves system maintainability, lowers power requirements, and shrinks the physical size. Furthermore, DolphinGemma’s predictive power integrated into CHAT could help identify mimics faster, making interactions more fluid and effective. Recognising that breakthroughs often stem from collaboration, Google intends to release DolphinGemma as an open model later this summer. While trained on Atlantic spotted dolphins, its architecture holds promise for researchers studying other cetaceans, potentially requiring fine-tuning for different species’ vocal repertoires.. The aim is to equip researchers globally with powerful tools to analyse their own acoustic datasets, accelerating the collective effort to understand these intelligent marine mammals. We are shifting from passive listening towards actively deciphering patterns, bringing the prospect of bridging the communication gap between our species perhaps just a little closer. See also: IEA: The opportunities and challenges of AI for global energy Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here. The post DolphinGemma: Google AI model understands dolphin chatter appeared first on AI News.

ChatGPT’s image generation feature has sparked a new wave of personalised digital creations, with LinkedIn users leading a trend of turning themselves into action figures. The craze began picking up momentum after the viral Studio Ghibli-style portraits sees users sharing images of themselves as boxed dolls – complete with accessories and job-themed packaging. There are several variations in the latest wave of AI-generated self-representation. The most common format is similar to a traditional action figure or Barbie doll, with props like coffee mugs, books, and laptops reflecting users’ professional lives. The images are designed to resemble toy store displays, complete with bold taglines and personalised packaging. The movement gained initial attention on LinkedIn, where professionals used the format to showcase their brand identities more playfully. The “AI Action Figure” format, in particular, resonated with marketers, consultants, and others looking to present themselves as standout figures – literally. Popularity of the service has since trickled into other platforms including Instagram, TikTok, and Facebook, though engagement remains largely centred around LinkedIn. ChatGPT’s image tool – part of its GPT-4o release – serves as the engine. Users upload a high-resolution photo of themselves, usually full-body, with a custom prompt describing how the final image should look. Details frequently include the person’s name, accessories, outfit styles, and package details. Some opt for a nostalgic “Barbiecore” vibe with pink tones and sparkles, while others stick to a corporate design that reflects their day job. Refinements are common. Many users go through multiple image generations, changing accessories and rewording prompts until the figure matches their wanted personality or profession. The result is a glossy, toy-style portrait that crosses the line between humour and personal branding. While the toy-style trend hasn’t seen the same viral reach as the Ghibli portrait craze, it has still sparked a steady flow of content across platforms. Hashtags like #AIBarbie and #BarbieBoxChallenge have gained traction, and some brands – including Mac Cosmetics and NYX – were quick to participate. A few public figures have joined in too, most notably US Representative Marjorie Taylor Greene, who shared a doll version of herself featuring accessories like a Bible and gavel. Regardless of the buzz, engagement levels are different. Many posts receive limited interaction, and most well-known influencers have avoided the trend. Nevertheless, it highlights ChatGPT’s growing presence in mainstream online culture, and its ability to respond to users’ creativity using relatively simple tools. The is not the first time ChatGPT’s image generation tool has overwhelmed the platform. When the Ghibli-style portraits first went viral, demand spiked so dramatically that OpenAI temporarily limited image generation for free accounts. CEO Sam Altman later described the surge in users as “biblical demand,” noting a dramatic rise in daily active users and infrastructure stress. The Barbie/action figure trend, though at a smaller scale, follows that same path – using ChatGPT’s simple interface and its growing popularity as a creative tool. As with other viral AI visuals, the trend has also raised broader conversations about identity, aesthetics, and self-presentation in digital spaces. However, unlike the Ghibli portrait craze, it hasn’t attracted much criticism – at least not yet. The format’s appeal lies in its simplicity. It offers users a way to engage with AI-generated art without needing technical skills, and satisfies an urge for of self-expression. The result is something like part professional head-shot, part novelty toy, and part visual joke, making it a surprisingly versatile format for social media sharing. While some may see the toy model phenomenon as a gimmick, others view it as a window into what’s possible when AI tools are placed directly in users’ hands. For now, whether it’s a mini-me holding a coffee mug or a Barbie-style figure ready for the toy shelf, ChatGPT is again changing how people choose to represent themselves in the digital age. (Photo by Unsplash) See also: ChatGPT hits record usage after viral Ghibli feature – Here are four risks to know first Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

A recent McKinsey report found that 75% of large enterprises are investing in digital twins to scale their AI solutions. Combining digital twins with AI has the potential to enhance the effectiveness of large language models and enable new applications for AI in real-time monitoring, offering significant business and operational benefits. What are digital twins? Digital twins, originally developed to aid in the design of complex machinery have evolved significantly over the last two decades. They track and analyse live systems in real-time by processing device telemetry, detecting shifting conditions, and enhancing situational awareness for operational managers. Powered by in-memory computing, they enable fast, actionable alerts. Beyond real-time monitoring, digital twins also can simulate intricate systems like those for use in airlines and logistics, supporting strategic planning and operational decisions through predictive analytics. Integrating digital twins with generative AI creates new opportunities for both technologies: The synergy can boost the prediction accuracy of generative AI, and can enhance the value of digital twins for system monitoring and development. Proactively identifying anomalies with AI-powered digital twins Continuous, real-time monitoring is a strategic necessity for organisations that manage complex live systems, like transportation networks, cybersecurity systems, and smart cities. Emerging problems must never be overlooked because delayed responses can cause small problems to become large ones. Enhancing digital twins with generative AI reshapes how real-time monitoring interprets massive volumes of live data, enabling the reliable and immediate detection of anomalies that impact operations. Generative AI can continuously examine analytics results produced by digital twins to uncover emerging trends and mitigate disruptions before they escalate. While AI enhances situational awareness for managers, it can also pinpoint new opportunities for optimising operations and boosting efficiency. At the same time, real-time data supplied by digital twins constrains the output of generative AI to avoid erratic results, like hallucinations. In a process called retrieval augmented generation, AI always uses the most up-to-date information about a live system to analyse behaviour and create recommendations. Transforming data interaction with AI-driven visualisations Unlocking insights from digital twin analytics should be intuitive, not technical. Generative AI is redefining how teams interact with massive datasets by enabling natural language-driven queries and visualisations. Instead of manually constructing intricate queries, users can simply describe their needs, and generative AI immediately visualises relevant charts and query results that provide new insights. This capability simplifies interactions and gives decision-makers the data they need. As organisations handle increasingly complex live systems, AI-powered intelligence allows them to efficiently sift through vast data pools, extract meaningful trends, and optimise operations with greater precision. It eliminates technical barriers, enabling faster, data-driven decisions that have a strategic impact. Incorporating machine learning with automatic retraining Digital twins can track numerous individual data streams and look for issues with the corresponding physical data sources. Working together, thousands or even millions of digital twins can monitor very large, complex systems. As messages flow in, each digital twin combines them with known information about a particular data source and analyses the data in a few milliseconds. It can incorporate a machine learning algorithm to assist in the analysis and find subtle issues that would be difficult to describe in hand-coded algorithms. After training with data from live operations, ML algorithms can identify anomalies and generate alerts for operational managers immediately. Once deployed to analyse live telemetry, an ML algorithm will likely encounter new situations not covered by its initial training set. It may either fail to detect anomalies or generate false positives. Automatic retraining lets the algorithm learn as it gains experience so it can improve its performance and adapt to changing conditions. Digital twins can work together to detect invalid ML responses and build new training sets that feed automatic retraining. By incorporating automatic retraining, businesses gain a competitive edge with real-time monitoring that reliably delivers actionable insights as it learns over time. Looking forward Integrating digital twin technology with generative AI and ML can transform how industries monitor complex, live systems by empowering better real-time insights and enabling managers to make faster, more informed decisions. ScaleOut Software’s newly-released Digital Twins™ Version 4 adds generative AI using OpenAI’s large language model and automatic ML retraining to move real-time monitoring towards the goal of fully-autonomous operations.  (Image source: Unsplash)

Generative AI is reshaping global competition and geopolitics, presenting challenges and opportunities for nations and businesses alike. Senior figures from Boston Consulting Group (BCG) and its tech division, BCG X, discussed the intricate dynamics of the global AI race, the dominance of superpowers like the US and China, the role of emerging “middle powers,” and the implications for multinational corporations. AI investments expose businesses to increasingly tense geopolitics Sylvain Duranton, Global Leader at BCG X, noted the significant geopolitical risk companies face: “For large companies, close to half of them, 44%, have teams around the world, not just in one country where their headquarters are.” Many of these businesses operate across numerous countries, making them vulnerable to differing regulations and sovereignty issues. “They’ve built their AI teams and ecosystem far before there was such tension around the world.” Duranton also pointed to the stark imbalance in the AI supply race, particularly in investment. Comparing the market capitalisation of tech companies, the US dwarfs Europe by a factor of 20 and the Asia Pacific region by five. Investment figures paint a similar picture, showing a “completely disproportionate” imbalance compared to the relative sizes of the economies. This AI race is fuelled by massive investments in compute power, frontier models, and the emergence of lighter, open-weight models changing the competitive dynamic.    Benchmarking national AI capabilities Nikolaus Lang, Global Leader at the BCG Henderson Institute – BCG’s think tank – detailed the extensive research undertaken to benchmark national GenAI capabilities objectively. The team analysed the “upstream of GenAI,” focusing on large language model (LLM) development and its six key enablers: capital, computing power, intellectual property, talent, data, and energy. Using hard data like AI researcher numbers, patents, data centre capacity, and VC investment, they created a comparative analysis. Unsurprisingly, the analysis revealed the US and China as the clear AI frontrunners and maintain leads in geopolitics. The US boasts the largest pool of AI specialists (around half a million), immense capital power ($303bn in VC funding, $212bn in tech R&D), and leading compute power (45 GW). Lang highlighted America’s historical dominance, noting, “the US has been the largest producer of notable AI models with 67%” since 1950, a lead reflected in today’s LLM landscape. This strength is reinforced by “outsized capital power” and strategic restrictions on advanced AI chip access through frameworks like the US AI Diffusion Framework.    China, the second AI superpower, shows particular strength in data—ranking highly in e-governance and mobile broadband subscriptions, alongside significant data centre capacity (20 GW) and capital power.  Despite restricted access to the latest chips, Chinese LLMs are rapidly closing the gap with US models. Lang mentioned the emergence of models like DeepSpeech as evidence of this trend, achieved with smaller teams, fewer GPU hours, and previous-generation chips. China’s progress is also fuelled by heavy investment in AI academic institutions (hosting 45 of the world’s top 100), a leading position in AI patent applications, and significant government-backed VC funding. Lang predicts “governments will play an important role in funding AI work going forward.” The middle powers: Europe, Middle East, and Asia Beyond the superpowers, several “middle powers” are carving out niches. EU: While trailing the US and China, the EU holds the third spot with significant data centre capacity (8 GW) and the world’s second-largest AI talent pool (275,000 specialists) when capabilities are combined. Europe also leads in top AI publications. Lang stressed the need for bundled capacities, suggesting AI, defence, and renewables are key areas for future EU momentum. Middle East (UAE & Saudi Arabia): These nations leverage strong capital power via sovereign wealth funds and competitively low electricity prices to attract talent and build compute power, aiming to become AI drivers “from scratch”. They show positive dynamics in attracting AI specialists and are climbing the ranks in AI publications.    Asia (Japan & South Korea): Leveraging strong existing tech ecosystems in hardware and gaming, these countries invest heavily in R&D (around $207bn combined by top tech firms). Government support, particularly in Japan, fosters both supply and demand. Local LLMs and strategic investments by companies like Samsung and SoftBank demonstrate significant activity.    Singapore: Singapore is boosting its AI ecosystem by focusing on talent upskilling programmes, supporting Southeast Asia’s first LLM, ensuring data centre capacity, and fostering adoption through initiatives like establishing AI centres of excellence.    The geopolitics of generative AI: Strategy and sovereignty The geopolitics of generative AI is being shaped by four clear dynamics: the US retains its lead, driven by an unrivalled tech ecosystem; China is rapidly closing the gap; middle powers face a strategic choice between building supply or accelerating adoption; and government funding is set to play a pivotal role, particularly as R&D costs climb and commoditisation sets in. As geopolitical tensions mount, businesses are likely to diversify their GenAI supply chains to spread risk. The race ahead will be defined by how nations and companies navigate the intersection of innovation, policy, and resilience. (Photo by Markus Krisetya) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

OpenAI has launched a legal counteroffensive against one of its co-founders, Elon Musk, and his competing AI venture, xAI. In court documents filed yesterday, OpenAI accuses Musk of orchestrating a “relentless” and “malicious” campaign designed to “take down OpenAI” after he left the organisation years ago. The court filing, submitted to the US District Court for the Northern District of California, alleges Musk could not tolerate OpenAI’s success after he had “abandoned and declared [it] doomed.” OpenAI is now seeking legal remedies, including an injunction to stop Musk’s alleged “unlawful and unfair action” and compensation for damages already caused.    Origin story of OpenAI and the departure of Elon Musk The legal documents recount OpenAI’s origins in 2015, stemming from an idea discussed by current CEO Sam Altman and President Greg Brockman to create an AI lab focused on developing artificial general intelligence (AGI) – AI capable of outperforming humans – for the “benefit of all humanity.” Musk was involved in the launch, serving on the initial non-profit board and pledging $1 billion in donations.    However, the relationship fractured. OpenAI claims that between 2017 and 2018, Musk’s demands for “absolute control” of the enterprise – or its potential absorption into Tesla – were rebuffed by Altman, Brockman, and then-Chief Scientist Ilya Sutskever. The filing quotes Sutskever warning Musk against creating an “AGI dictatorship.” Following this disagreement, OpenAI alleges Elon Musk quit in February 2018, declaring the venture would fail without him and that he would pursue AGI development at Tesla instead. Critically, OpenAI contends the pledged $1 billion “was never satisfied—not even close”.    Restructuring, success, and Musk’s alleged ‘malicious’ campaign Facing escalating costs for computing power and talent retention, OpenAI restructured and created a “capped-profit” entity in 2019 to attract investment while remaining controlled by the non-profit board and bound by its mission. This structure, OpenAI states, was announced publicly and Musk was offered equity in the new entity but declined and raised no objection at the time.    OpenAI highlights its subsequent breakthroughs – including GPT-3, ChatGPT, and GPT-4 – achieved massive public adoption and critical acclaim. These successes, OpenAI emphasises, were made after the departure of Elon Musk and allegedly spurred his antagonism. The filing details a chronology of alleged actions by Elon Musk aimed at harming OpenAI:    Founding xAI: Musk “quietly created” his competitor, xAI, in March 2023.    Moratorium call: Days later, Musk supported a call for a development moratorium on AI more advanced than GPT-4, a move OpenAI claims was intended “to stall OpenAI while all others, most notably Musk, caught up”.    Records demand: Musk allegedly made a “pretextual demand” for confidential OpenAI documents, feigning concern while secretly building xAI.    Public attacks: Using his social media platform X (formerly Twitter), Musk allegedly broadcast “press attacks” and “malicious campaigns” to his vast following, labelling OpenAI a “lie,” “evil,” and a “total scam”.    Legal actions: Musk filed lawsuits, first in state court (later withdrawn) and then the current federal action, based on what OpenAI dismisses as meritless claims of a “Founding Agreement” breach.    Regulatory pressure: Musk allegedly urged state Attorneys General to investigate OpenAI and force an asset auction.    “Sham bid”: In February 2025, a Musk-led consortium made a purported $97.375 billion offer for OpenAI, Inc.’s assets. OpenAI derides this as a “sham bid” and a “stunt” lacking evidence of financing and designed purely to disrupt OpenAI’s operations, potential restructuring, fundraising, and relationships with investors and employees, particularly as OpenAI considers evolving its capped-profit arm into a Public Benefit Corporation (PBC). One investor involved allegedly admitted the bid’s aim was to gain “discovery”.    Based on these allegations, OpenAI asserts two primary counterclaims against both Elon Musk and xAI: Unfair competition: Alleging the “sham bid” constitutes an unfair and fraudulent business practice under California law, intended to disrupt OpenAI and gain an unfair advantage for xAI.    Tortious interference with prospective economic advantage: Claiming the sham bid intentionally disrupted OpenAI’s existing and potential relationships with investors, employees, and customers.  OpenAI argues Musk’s actions have forced it to divert resources and expend funds, causing harm. They claim his campaign threatens “irreparable harm” to their mission, governance, and crucial business relationships. The filing also touches upon concerns regarding xAI’s own safety record, citing reports of its AI Grok generating harmful content and misinformation. The counterclaims mark a dramatic escalation in the legal battle between the AI pioneer and its departed co-founder. While Elon Musk initially sued OpenAI alleging a betrayal of its founding non-profit, open-source principles, OpenAI now contends Musk’s actions are a self-serving attempt to undermine a competitor he couldn’t control. With billions at stake and the future direction of AGI in the balance, this dispute is far from over. Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The International Energy Agency (IEA) has explored the opportunities and challenges brought about by AI with regards to global energy.   Training and deploying sophisticated AI models occur within vast, power-hungry data centres. A “typical AI-focused data centre consumes as much electricity as 100 000 households,” the IEA notes, with the largest facilities under construction projected to demand 20x times that amount. Surging data centre investments Global investment in data centres has nearly doubled since 2022, reaching half a trillion dollars in 2024, sparking concerns about escalating electricity needs. While data centres accounted for approximately 1.5% of global electricity consumption in 2024 (around 415 terawatt-hours, TWh,) their local impact is far more significant. Consumption has grown annually by about 12% since 2017, vastly outpacing overall electricity demand growth. The US leads this consumption (45%), followed by China (25%) and Europe (15%). Almost half of US data centre capacity is concentrated in just five regional clusters. Looking ahead, the IEA projects global data centre electricity consumption to more than double by 2030 to reach approximately 945 TWh. To put that in context, that’s slightly more than Japan’s current total electricity consumption. AI is pinpointed as the “most important driver of this growth”. The US is projected to see the largest increase, where data centres could account for nearly half of all electricity demand growth by 2030. By the decade’s end, US data centres are forecast to consume more electricity than the combined usage of its aluminium, steel, cement, chemical, and other energy-intensive manufacturing industries. The IEA’s “Base Case” extends this trajectory, anticipating around 1,200 TWh of global data centre electricity consumption by 2035. However, significant uncertainties exist, with projections for 2035 ranging from 700 TWh (“Headwinds Case”) to 1,700 TWh (“Lift-Off Case”) depending on AI uptake, efficiency gains, and energy sector bottlenecks. Fatih Birol, Executive Director of the IEA, said: “AI is one of the biggest stories in the energy world today – but until now, policymakers and markets lacked the tools to fully understand the wide-ranging impacts. “In the United States, data centres are on course to account for almost half of the growth in electricity demand; in Japan, more than half; and in Malaysia, as much as one-fifth.” Meeting the global AI energy demand Powering this AI boom requires a diverse energy portfolio. The IEA suggests renewables and natural gas will take the lead, but emerging technologies like small modular nuclear reactors (SMRs) and advanced geothermal also have a role. Renewables, supported by storage and grid infrastructure, are projected to meet half the growth in data centre demand globally up to 2035. Natural gas is also crucial, particularly in the US, expanding by 175 TWh to meet data centre needs by 2035 in the Base Case. Nuclear power contributes similarly, especially in China, Japan, and the US, with the first SMRs expected around 2030. However, simply increasing generation isn’t sufficient. The IEA stresses the critical need for infrastructure upgrades, particularly grid investment. Existing grids are already strained, potentially delaying around 20% of planned data centre projects globally due to complex connection queues and long lead times for essential components like transformers. The potential of AI to optimise energy systems Beyond its energy demands, AI offers significant potential to revolutionise the energy sector itself. The IEA details numerous applications: Energy supply: The oil and gas industry – an early adopter – uses AI to optimise exploration, production, maintenance, and safety, including reducing methane emissions. AI can also aid critical mineral exploration. Electricity sector: AI can improve forecasting for variable renewables, reducing curtailment. It enhances grid balancing, fault detection (reducing outage durations by 30-50%), and can unlock significant transmission capacity through smarter management—potentially 175 GW without building new lines. End uses: In industry, widespread AI adoption for process optimisation could yield energy savings equivalent to Mexico’s total energy consumption today. Transport applications like traffic management and route optimisation could save energy equivalent to 120 million cars, though rebound effects from autonomous vehicles need monitoring. Building optimisation potential is significant but hampered by slower digitalisation. Innovation: AI can dramatically accelerate the discovery and testing of new energy technologies, such as advanced battery chemistries, catalysts for synthetic fuels, and carbon capture materials. However, the energy sector currently underutilises AI for innovation compared to fields like biomedicine. Collaboration is key to navigating challenges Despite the potential, significant barriers hinder AI’s full integration into the energy sector. These include data access and quality issues, inadequate digital infrastructure and skills (AI talent concentration is lower in energy sectors,) regulatory hurdles, and security concerns. Cybersecurity is a double-edged sword: while AI enhances defence capabilities, it also equips attackers with sophisticated tools. Cyberattacks on utilities have tripled in the last four years. Supply chain security is another critical concern, particularly regarding critical minerals like gallium (used in advanced chips,) where supply is highly concentrated. The IEA concludes that deeper dialogue and collaboration between the technology sector, the energy industry, and policymakers are paramount. Addressing grid integration challenges requires smarter data centre siting, exploring operational flexibility, and streamlining permitting. While AI presents opportunities for substantial emissions reductions through optimisation, exceeding the emissions generated by data centres, these gains are not guaranteed and could be offset by rebound effects. “AI is a tool, potentially an incredibly powerful one, but it is up to us – our societies, governments, and companies – how we use it,” said Dr Birol. “The IEA will continue to provide the data, analysis, and forums for dialogue to help policymakers and other stakeholders navigate the path ahead as the energy sector shapes the future of AI, and AI shapes the future of energy.” (Photo by Javier Miranda) See also: UK forms AI Energy Council to align growth and sustainability goals Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here. The post IEA: The opportunities and challenges of AI for global energy appeared first on AI News.

Nina Schick is a leading speaker and expert on generative AI, renowned for her groundbreaking work at the intersection of technology, society and geopolitics. As one of the first authors to publish a book on generative AI, she has emerged as a sought-after speaker helping global leaders, businesses, and institutions understand and adapt to this transformative moment. We spoke to Nina to explore the future of AI-driven innovation, its ethical and political dimensions, and how organisations can lead in this rapidly evolving landscape. In your view, how will generative AI redefine the foundational structures of business and economic productivity in the coming decade? I believe generative AI is absolutely going to transform the entire economy as we know it. This moment feels quite similar to around 1993, when we were first being told to prepare for the Internet. Back then, some thirty years ago, we didn’t fully grasp, in our naivety, how profoundly the Internet would go on to reshape business and the broader global economy. Now, we are witnessing something even more significant. You can think of generative AI as a kind of new combustion engine, but for all forms of human creative and intelligent activity. It’s a fundamental enabler. Every industry, every facet of productivity, will be impacted and ultimately transformed by generative AI. We’re already beginning to see those use cases emerge, and this is only the beginning. As AI and data continue to evolve as forces shaping society, how do you see them redefining the political agenda and global power dynamics? When you reflect on just how profound AI is in its capacity to reshape the entire framework of society, it becomes clear that this AI revolution is going to emerge as one of the most important political questions of our generation. Over the past 30 years, we’ve already seen how the information revolution — driven by the Internet, smartphones, and cloud computing — has become a defining geopolitical force. Now, we’re layering the AI revolution on top of that, along with the data that fuels it, and the impact is nothing short of seismic. This will evolve into one of the most pressing and influential issues society must address over the coming decades. So, to answer the question directly — AI won’t just influence politics; it will, in many ways, become the very fabric of politics itself. There’s been much discussion about the Metaverse and immersive tech — how do you see these experiences evolving, and what role do you believe AI will play in architecting this next frontier of digital interaction? The Metaverse represents a vision for where the Internet may be heading — a future where digital experiences become far more immersive, intuitive, and experiential. It’s a concept that imagines how we might engage with digital content in a far more lifelike way. But the really fascinating element here is that artificial intelligence is the key enabler — the actual vehicle — that will allow us to build and scale these kinds of immersive digital environments. So, even though the Metaverse remains largely an untested concept in terms of its final form, what is clear right now is that AI is going to be the engine that generates and populates the content that will live within these immersive spaces. Considering the transformative power of AI and big data, what ethical imperatives must policymakers and society address to ensure equitable and responsible deployment? The conversation around ethics, artificial intelligence, and big data is one that is set to become intensely political and highly consequential. It will likely remain a predominant issue for many years to come. What we’re dealing with here is a technology so transformative that it has the potential to reshape the economy, redefine the labour market, and fundamentally alter the structure of society itself. That’s why the ethical questions — how to ensure this technology is applied in a fair, safe, and responsible manner — will be one of the defining political challenges of our time. For business leaders navigating digital transformation, what mindset shifts are essential to meaningfully integrate AI into long-term strategy and operations? For businesses aiming to digitally transform, especially in the era of artificial intelligence, it’s critical to first understand the conceptual paradigm shift we are currently undergoing. Once that foundational understanding is in place, it becomes much easier to explore and adopt AI technologies effectively. If companies wish to remain competitive and gain a strategic edge, now is the time to start investigating how generative AI can be thoughtfully and effectively integrated into their business models. This includes identifying priority areas where AI can deliver long-term value — not just short-term. If you put together a generative AI working group to look into this, your business will be transformed and able to compete with other businesses that are using AI to transform their processes. As one of the earliest voices to articulate the societal implications of generative AI, what catalysed your foresight to explore this space before it entered the mainstream conversation? My interest in AI didn’t come from a technical background. I’m not a techie. My experience has always been in analysing macro trends that shape society, geopolitics, and the wider world. That perspective is what led me to AI, as it quickly became clear that this technology would have far-reaching societal implications. I began researching and writing about AI because I saw it as more than just a technological shift. Ultimately, this isn’t only a story about innovation. It’s a story about humanity. Generative AI, as an exponential technology built and directed by humans, is going to transform not just the way we work, but the way we live. It will even challenge our understanding of what it means to be human. Photo by Heidi Fin on Unsplash Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. The post Nina Schick, author: Generative AI’s impact on business, politics and society appeared first on AI News.

Spot AI has introduced Iris, which the company describes as the world’s first universal video AI agent builder for enterprise camera systems. The tool allows businesses to create customised AI agents through a conversational interface, making it easier to monitor and act on video data from physical settings without the need for technical expertise. Designed for industries like manufacturing, logistics, retail, construction, and healthcare, Iris builds on Spot AI’s earlier launch of out-of-the-box Video AI Agents for safety, security, and operations. While those prebuilt agents focus on common use cases, Iris gives organisations the flexibility to train agents for more specific, business-critical scenarios. According to Spot AI, users can build video agents in a matter of minutes. The system allows training through reinforcement—using examples of what the AI should and shouldn’t detect—and can be configured to trigger real-world responses like shutting down equipment, locking doors, or generating alerts. CEO and Co-Founder Rish Gupta said the tool dramatically shortens the time required to create specialised video detection systems. “What used to take months of development now happens in minutes,” Gupta explained. Before Iris, creating specialised video detection required dedicated AI/ML teams with advanced degrees, thousands of annotated images, and 8 weeks of complex development,” he explained. “Iris puts that same power in the hands of any business leader through simple conversation with 8 minutes and 20 training images.” Examples from real-world settings Spot AI highlighted a variety of industry-specific use cases that Iris could support: Manufacturing: Detecting product backups or fluid leaks, with automatic responses based on severity. Warehousing: Spotting unsafe stacking of boxes or pallets to prevent accidents. Retail: Monitoring shelf stock levels and generating alerts for restocking. Healthcare: Distinguishing between staff and patients wearing similar uniforms to optimise traffic flow and safety. Security: Identifying tools like bolt cutters in parking areas to address evolving security threats. Safety compliance: Verifying whether workers are wearing required safety gear on-site. Video AI agents continuously monitor critical areas and help teams respond quickly to safety hazards, operational inefficiencies, and security issues. With Iris, those agents can be developed and modified through natural language interaction, reducing the need for engineering support and making video insights more accessible across departments. Iris is part of Spot AI’s broader effort to make video data more actionable in physical environments. The company plans to discuss the tool and its capabilities at Google Cloud Next, where Rish Gupta is scheduled to speak during a media roundtable on April 9. (Image by Spot AI) See also: ChatGPT hits record usage after viral Ghibli feature—Here are four risks to know first Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The promise of AI is that it’ll make all of our lives easier. And with great convenience comes the potential for serious profit. The United Nations thinks AI could be a $4.8 trillion global market by 2033 – about as big as the German economy. But forget about 2033: in the here and now, AI is already fueling transformation in industries as diverse as financial services, manufacturing, healthcare, marketing, agriculture, and e-commerce. Whether it’s autonomous algorithmic ‘agents’ managing your investment portfolio or AI diagnostics systems detecting diseases early, AI is fundamentally changing how we live and work. But cynicism is snowballing around AI – we’ve seen Terminator 2 enough times to be extremely wary. The question worth asking, then, is how do we ensure trust as AI integrates deeper into our everyday lives? The stakes are high: A recent report by Camunda highlights an inconvenient truth: most organisations (84%) attribute regulatory compliance issues to a lack of transparency in AI applications. If companies can’t view algorithms – or worse, if the algorithms are hiding something – users are left completely in the dark. Add the factors of systemic bias, untested systems, and a patchwork of regulations and you have a recipe for mistrust on a large scale. Transparency: Opening the AI black box For all their impressive capabilities, AI algorithms are often opaque, leaving users ignorant of how decisions are reached. Is that AI-powered loan request being denied because of your credit score – or due to an undisclosed company bias? Without transparency, AI can pursue its owner’s goals, or that of its owner, while the user remains unaware, still believing it’s doing their bidding. One promising solution would be to put the processes on the blockchain, making algorithms verifiable and auditable by anyone. This is where Web3 tech comes in. We’re already seeing startups explore the possibilities. Space and Time (SxT), an outfit backed by Microsoft, offers tamper-proof data feeds consisting of a verifiable compute layer, so SxT can ensure that the information on which AI relies is real, accurate, and untainted by a single entity. Space and Time’s novel Proof of SQL prover guarantees queries are computed accurately against untampered data, proving computations in blockchain histories and being able to do so much faster than state-of-the art zkVMs and coprocessors. In essence, SxT helps establish trust in AI’s inputs without dependence on a centralised power. Proving AI can be trusted Trust isn’t a one-and-done deal; it’s earned over time, analogous to a restaurant maintaining standards to retain its Michelin star. AI systems must be assessed continually for performance and safety, especially in high-stakes domains like healthcare or autonomous driving. A second-rate AI prescribing the wrong medicines or hitting a pedestrian is more than a glitch, it’s a catastrophe. This is the beauty of open-source models and on-chain verification via using immutable ledgers, with built-in privacy protections assured by the use of cryptography like Zero-Knowledge Proofs (ZKPs). Trust isn’t the only consideration, however: Users must know what AI can and can’t do, to set their expectations realistically. If a user believes AI is infallible, they’re more likely to trust flawed output. To date, the AI education narrative has centred on its dangers. From now on, we should try to improve users’ knowledge of AI’s capabilities and limitations, better to ensure users are empowered not exploited. Compliance and accountability As with cryptocurrency, the word compliance comes often when discussing AI. AI doesn’t get a pass under the law and various regulations. How should a faceless algorithm be held accountable? The answer may lie in the modular blockchain protocol Cartesi, which ensures AI inference happens on-chain. Cartesi’s virtual machine lets developers run standard AI libraries – like TensorFlow, PyTorch, and Llama.cpp – in a decentralised execution environment, making it suitable for on-chain AI development. In other words, a blend of blockchain transparency and computational AI. Trust through decentralisation The UN’s recent Technology and Innovation Report shows that while AI promises prosperity and innovation, its development risks “deepening global divides.” Decentralisation could be the answer, one that helps AI scale and instils trust in what’s under the hood. (Image source: Unsplash)

AI is driving a new revolution across a number of industries and the supply chain is no exception. AI has been the most transformative technology of the decade, and it’s no secret it has helped supply chains become more efficient, resilient, and responsive, while allowing organisations to become more efficient and ensuring workforces to focus on more strategic growth. However despite the benefits of the technology, many businesses are slow to adopt the technology, with recent statistics showing only one in ten of SME’s regularly use AI technology, indicating companies and employees are still not operating at their full potential, thus missing out on opportunities for growth and optimisation.  Transforming the supply chain through AI The potential that AI has in the supply chain is undeniable, with some estimating that AI helps businesses reduce logistics costs by 15%, reduce inventory levels by 35% and raise service levels by 65%. In contrast, failure to implement AI tools could set companies back, leave employees feeling unmotivated and unproductive and result in a weak supply chain and poor staff retention. Now, more than ever, it’s time for businesses to not just pay lip service to AI – they must start using it within their supply chains to truly enhance operations. Due to the evolving market dynamics, AI is not just a competitive advantage; it’s essential for business agility and profitability. Here are two ways in which organisations can use AI to improve their supply chains. Automating the supply chain & harnessing the power of AI for resilience AI allows businesses to tackle supply chain challenges head-on by automating time-consuming manual processes, such as data-logging whilst reducing errors. By taking over repetitive and potentially hazardous tasks, AI frees up employees to focus on strategic initiatives that drive business value. For example, a recent report highlighted that nearly three quarters of warehouse staff surveyed are excited about the possibilities of generative AI and robotics improving their job roles. Needless to say, a supply chain still can’t operate at its peak without resilience – which is the capacity of a supply chain to withstand and recover from disruptions – ensuring uninterrupted operations and minimal impact to businesses and customers. As global markets continue to evolve & expand, businesses are challenged to adapt swiftly to unforeseen disruptions. AI enables businesses to provide real time data analysis, providing unprecedented insights into the web of supply chain dynamics and acting as the eyes and ears of a supply chain. This empowers each component with the ability to make informed decisions quickly to meet supply chain demands. Allowing insights into every aspect of their warehouse operations, real time data enables visibility which permits precise monitoring, enhanced customer service and reduced downtime – identifying potential issues before they become a major problem. At the heart of the supply chain is communication between all stakeholders, with technology such as AI providing real time data, seamless collaboration is enabled by providing a shared platform where suppliers, manufacturers, and distributors can exchange information instantaneously. Enhanced communication leads to quicker issue resolution, enabling the supply chain to adapt rapidly to changing circumstances. Robotics, AI and real-time data introduce an all-encompassing visibility of the good’s journey, which leads to resilience. Human expertise with robot precision Building on the theme of resilience, in the next couple of years the industry will witness AI-integrated robots becoming collaborative partners to their human co-workers. Particularly in environments requiring vast coverage and extensive data capture, robots that are equipped with groundbreaking sensor technologies will navigate, adapt and work with greater levels of autonomy along with other machinery and people in busy environments. This will result in speed of data acquisition and most importantly, allowing companies to make decisions based on actionable insights a lot faster than ever before.  These advancements will transform robots into true cobots and will take human-robot teamwork to an unprecedented level. We will also see that robots will become better with understanding nuanced human gestures and intentions. This evolution in collaboration with technology will redefine what humans and machines can accomplish together. What’s next for the industry? In theory implementing AI and advanced technology in the supply chain has the potential to bring significant benefits. However, we will only begin to see substantial results once these innovations are widely adopted in practice. By automating the supply chain and using data to fuel predictions, these technologies are the foundations for a new industrial revolution that will shape the future of the industries for years to come. Those that delay starting their journeys will risk being left behind. Photo by Miltiadis Fragkidis on Unsplash Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Chinese AI startup DeepSeek has solved a problem that has frustrated AI researchers for several years. Its breakthrough in AI reward models could improve dramatically how AI systems reason and respond to questions. In partnership with Tsinghua University researchers, DeepSeek has created a technique detailed in a research paper, titled “Inference-Time Scaling for Generalist Reward Modeling.” It outlines how a new approach outperforms existing methods and how the team “achieved competitive performance” compared to strong public reward models. The innovation focuses on enhancing how AI systems learn from human preferences – a important aspect of creating more useful and aligned artificial intelligence. What are AI reward models, and why do they matter? AI reward models are important components in reinforcement learning for large language models. They provide feedback signals that help guide an AI’s behaviour toward preferred outcomes. In simpler terms, reward models are like digital teachers that help AI understand what humans want from their responses. “Reward modeling is a process that guides an LLM towards human preferences,” the DeepSeek paper states. Reward modeling becomes important as AI systems get more sophisticated and are deployed in scenarios beyond simple question-answering tasks. The innovation from DeepSeek addresses the challenge of obtaining accurate reward signals for LLMs in different domains. While current reward models work well for verifiable questions or artificial rules, they struggle in general domains where criteria are more diverse and complex. The dual approach: How DeepSeek’s method works DeepSeek’s approach combines two methods: Generative reward modeling (GRM): This approach enables flexibility in different input types and allows for scaling during inference time. Unlike previous scalar or semi-scalar approaches, GRM provides a richer representation of rewards through language. Self-principled critique tuning (SPCT): A learning method that fosters scalable reward-generation behaviours in GRMs through online reinforcement learning, one that generates principles adaptively. One of the paper’s authors from Tsinghua University and DeepSeek-AI, Zijun Liu, explained that the combination of methods allows “principles to be generated based on the input query and responses, adaptively aligning reward generation process.” The approach is particularly valuable for its potential for “inference-time scaling” – improving performance by increasing computational resources during inference rather than just during training. The researchers found that their methods could achieve better results with increased sampling, letting models generate better rewards with more computing. Implications for the AI Industry DeepSeek’s innovation comes at an important time in AI development. The paper states “reinforcement learning (RL) has been widely adopted in post-training for large language models […] at scale,” leading to “remarkable improvements in human value alignment, long-term reasoning, and environment adaptation for LLMs.” The new approach to reward modelling could have several implications: More accurate AI feedback: By creating better reward models, AI systems can receive more precise feedback about their outputs, leading to improved responses over time. Increased adaptability: The ability to scale model performance during inference means AI systems can adapt to different computational constraints and requirements. Broader application: Systems can perform better in a broader range of tasks by improving reward modelling for general domains. More efficient resource use: The research shows that inference-time scaling with DeepSeek’s method could outperform model size scaling in training time, potentially allowing smaller models to perform comparably to larger ones with appropriate inference-time resources. DeepSeek’s growing influence The latest development adds to DeepSeek’s rising profile in global AI. Founded in 2023 by entrepreneur Liang Wenfeng, the Hangzhou-based company has made waves with its V3 foundation and R1 reasoning models. The company upgraded its V3 model (DeepSeek-V3-0324) recently, which the company said offered “enhanced reasoning capabilities, optimised front-end web development and upgraded Chinese writing proficiency.” DeepSeek has committed to open-source AI, releasing five code repositories in February that allow developers to review and contribute to development. While speculation continues about the potential release of DeepSeek-R2 (the successor to R1) – Reuters has speculated on possible release dates – DeepSeek has not commented in its official channels. What’s next for AI reward models? According to the researchers, DeepSeek intends to make the GRM models open-source, although no specific timeline has been provided. Open-sourcing will accelerate progress in the field by allowing broader experimentation with reward models. As reinforcement learning continues to play an important role in AI development, advances in reward modelling like those in DeepSeek and Tsinghua University’s work will likely have an impact on the abilities and behaviour of AI systems. Work on AI reward models demonstrates that innovations in how and when models learn can be as important increasing their size. By focusing on feedback quality and scalability, DeepSeek addresses one of the fundamental challenges to creating AI that understands and aligns with human preferences better. See also: DeepSeek disruption: Chinese AI innovation narrows global technology divide Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Deep Cogito has released several open large language models (LLMs) that outperform competitors and claim to represent a step towards achieving general superintelligence. The San Francisco-based company, which states its mission is “building general superintelligence,” has launched preview versions of LLMs in 3B, 8B, 14B, 32B, and 70B parameter sizes. Deep Cogito asserts that “each model outperforms the best available open models of the same size, including counterparts from LLAMA, DeepSeek, and Qwen, across most standard benchmarks”. Impressively, the 70B model from Deep Cogito even surpasses the performance of the recently released Llama 4 109B Mixture-of-Experts (MoE) model.    Iterated Distillation and Amplification (IDA) Central to this release is a novel training methodology called Iterated Distillation and Amplification (IDA).  Deep Cogito describes IDA as “a scalable and efficient alignment strategy for general superintelligence using iterative self-improvement”. This technique aims to overcome the inherent limitations of current LLM training paradigms, where model intelligence is often capped by the capabilities of larger “overseer” models or human curators. The IDA process involves two key steps iterated repeatedly: Amplification: Using more computation to enable the model to derive better solutions or capabilities, akin to advanced reasoning techniques. Distillation: Internalising these amplified capabilities back into the model’s parameters. Deep Cogito says this creates a “positive feedback loop” where model intelligence scales more directly with computational resources and the efficiency of the IDA process, rather than being strictly bounded by overseer intelligence. “When we study superintelligent systems,” the research notes, referencing successes like AlphaGo, “we find two key ingredients enabled this breakthrough: Advanced Reasoning and Iterative Self-Improvement”. IDA is presented as a way to integrate both into LLM training. Deep Cogito claims IDA is efficient, stating the new models were developed by a small team in approximately 75 days. They also highlight IDA’s potential scalability compared to methods like Reinforcement Learning from Human Feedback (RLHF) or standard distillation from larger models. As evidence, the company points to their 70B model outperforming Llama 3.3 70B (distilled from a 405B model) and Llama 4 Scout 109B (distilled from a 2T parameter model). Capabilities and performance of Deep Cogito models The newly released Cogito models – based on Llama and Qwen checkpoints – are optimised for coding, function calling, and agentic use cases. A key feature is their dual functionality: “Each model can answer directly (standard LLM), or self-reflect before answering (like reasoning models),” similar to capabilities seen in models like Claude 3.5. However, Deep Cogito notes they “have not optimised for very long reasoning chains,” citing user preference for faster answers and the efficiency of distilling shorter chains. Extensive benchmark results are provided, comparing Cogito models against size-equivalent state-of-the-art open models in both direct (standard) and reasoning modes. Across various benchmarks (MMLU, MMLU-Pro, ARC, GSM8K, MATH, etc.) and model sizes (3B, 8B, 14B, 32B, 70B,) the Cogito models generally show significant performance gains over counterparts like Llama 3.1/3.2/3.3 and Qwen 2.5, particularly in reasoning mode. For instance, the Cogito 70B model achieves 91.73% on MMLU in standard mode (+6.40% vs Llama 3.3 70B) and 91.00% in thinking mode (+4.40% vs Deepseek R1 Distill 70B). Livebench scores also show improvements. Here are benchmarks of 14B models for a medium-sized comparison: While acknowledging benchmarks don’t fully capture real-world utility, Deep Cogito expresses confidence in practical performance. This release is labelled a preview, with Deep Cogito stating they are “still in the early stages of this scaling curve”. They plan to release improved checkpoints for the current sizes and introduce larger MoE models (109B, 400B, 671B) “in the coming weeks / months”. All future models will also be open-source. (Photo by Pietro Mattia) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Alibaba Cloud has expanded its AI portfolio for global customers with a raft of new models, platform enhancements, and Software-as-a-Service (SaaS) tools. The announcements, made during its Spring Launch 2025 online event, underscore the drive by Alibaba to accelerate AI innovation and adoption on a global scale. The digital technology and intelligence arm of Alibaba is focusing on meeting increasing demand for AI-driven digital transformation worldwide. Selina Yuan, President of International Business at Alibaba Cloud Intelligence, said: “We are launching a series of Platform-as-a-Service（PaaS) and AI capability updates to meet the growing demand for digital transformation from across the globe. “These upgrades allow us to deliver even more secure and high-performance services that empower businesses to scale and innovate in an AI-driven world.” Alibaba expands access to foundational AI models Central to the announcement is the broadened availability of Alibaba Cloud’s proprietary Qwen large language model (LLM) series for international clients, initially accessible via its Singapore availability zones. This includes several specialised models: Qwen-Max: A large-scale Mixture of Experts (MoE) model. QwQ-Plus: An advanced reasoning model designed for complex analytical tasks, sophisticated question answering, and expert-level mathematical problem-solving. QVQ-Max: A visual reasoning model capable of handling complex multimodal problems, supporting visual input and chain-of-thought output for enhanced accuracy. Qwen2.5-Omni-7b: An end-to-end multimodal model. These additions provide international businesses with more powerful and diverse tools for developing sophisticated AI applications. Platform enhancements power AI scale To support these advanced models, Alibaba Cloud’s Platform for AI (PAI) received significant upgrades aimed at delivering scalable, cost-effective, and user-friendly generative AI solutions. Key enhancements include the introduction of distributed inference capabilities within the PAI-Elastic Algorithm Service (EAS). Utilising a multi-node architecture, this addresses the computational demands of super-large models – particularly those employing MoE structures or requiring ultra-long-text processing – to overcome limitations inherent in traditional single-node setups. Furthermore, PAI-EAS now features a prefill-decode disaggregation function designed to boost performance and reduce operational costs. Alibaba Cloud reported impressive results when deploying this with the Qwen2.5-72B model, achieving a 92% increase in concurrency and a 91% boost in tokens per second (TPS). The PAI-Model Gallery has also been refreshed, now offering nearly 300 open-source models—including the complete range of Alibaba Cloud’s own open-source Qwen and Wan series. These are accessible via a no-code deployment and management interface. Additional new PAI-Model Gallery features – like model evaluation and model distillation (transferring knowledge from large to smaller, more cost-effective models) – further enhance its utility. Alibaba integrates AI into data management Alibaba Cloud’s flagship cloud-native relational database, PolarDB, now incorporates native AI inference powered by Qwen. PolarDB’s in-database machine learning capability eliminates the need to move data for inference workflows, which significantly cuts processing latency while improving efficiency and data security. The feature is optimised for text-centric tasks such as developing conversational Retrieval-Augmented Generation (RAG) agents, generating text embeddings, and performing semantic similarity searches. Additionally, the company’s data warehouse, AnalyticDB, is now integrated into Alibaba Cloud’s generative AI development platform Model Studio. This integration serves as the recommended vector database for RAG solutions. This allows organisations to connect their proprietary knowledge bases directly with AI models on the platform to streamline the creation of context-aware applications. Beyond infrastructure and platform layers, Alibaba Cloud introduced two new SaaS AI tools: AI Doc: An intelligent document processing tool using LLMs to parse diverse documents (reports, forms, manuals) efficiently. It extracts specific information and can generate tailored reports, such as ESG reports when integrated with Alibaba Cloud’s Energy Expert sustainability solution. Smart Studio: An AI-powered content creation platform supporting text-to-image, image-to-image, and text-to-video generation. It aims to enhance marketing and creative outputs in sectors like e-commerce, gaming, and entertainment, enabling features like virtual try-ons or generating visuals from text descriptions. All these developments follow Alibaba’s announcement in February of a $53 billion investment over the next three years dedicated to advancing its cloud computing and AI infrastructure. This colossal investment, noted as exceeding the company’s total AI and cloud expenditure over the previous decade, highlights a deep commitment to AI-driven growth and solidifies its position as a major global cloud provider. “As cloud and AI become essential for global growth, we are committed to enhancing our core product offerings to address our customers’ evolving needs,” concludes Yuan. Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The UK government has announced the first meeting of a new AI Energy Council aimed at ensuring the nation’s AI and clean energy goals work in tandem to drive economic growth. The inaugural meeting of the council will see members agree on its core objectives, with a central focus on how the government’s mission to become a clean energy superpower can support its commitment to advancing AI and compute infrastructure. Unveiled earlier this year as part of the government’s response to the AI Opportunities Action Plan, the council will serve as a crucial platform for bringing together expert insights on the significant energy demands associated with the AI sector. Concerns surrounding the substantial energy requirements of AI data centres are a global challenge. The UK is proactively addressing this issue through initiatives like the establishment of new AI Growth Zones. These zones are dedicated hubs for AI development that are strategically located in areas with access to at least 500MW of power—an amount equivalent to powering approximately two million homes. This approach is designed to attract private investment from companies looking to establish operations in Britain, ultimately generating local jobs and boosting the economy. Peter Kyle, Secretary of State for Science, Innovation, and Technology, said: “The work of the AI Energy Council will ensure we aren’t just powering our AI needs to deliver new waves of opportunity in all parts of the country, but can do so in a way which is responsible and sustainable. “This requires a broad range of expertise from industry and regulators as we fire up the UK’s economic engine to make it fit for the age of AI—meaning we can deliver the growth which is the beating heart of our Plan for Change.” The Council is also expected to delve into the role of clean energy sources, including renewables and nuclear, in powering the AI revolution. A key aspect of its work will involve advising on how to improve energy efficiency and sustainability within AI and data centre infrastructure, with specific considerations for resource usage such as water. Furthermore, the council will take proactive steps to ensure the secure adoption of AI across the UK’s critical energy network itself. Ed Miliband, Secretary of State for Energy Security and Net Zero, commented: “We are making the UK a clean energy superpower, building the homegrown energy this country needs to protect consumers and businesses, and drive economic growth, as part of our Plan for Change. “AI can play an important role in building a new era of clean electricity for our country and as we unlock AI’s potential, this Council will help secure a sustainable scale up to benefit businesses and communities across the UK.” In a parallel effort to facilitate the growth of the AI sector, the UK government has been working closely with energy regulator Ofgem and the National Energy System Operator (NESO) to implement fundamental reforms to the UK’s connections process. Subject to final sign-offs from Ofgem, these reforms could potentially unlock more than 400GW of capacity from the connection queue. This acceleration of projects is deemed vital for economic growth, particularly for the delivery of new large-scale AI data centres that require significant power infrastructure. The newly-formed AI Energy Council comprises representatives from 14 key organisations across the energy and technology sectors, including regulators and leading companies. These members will contribute their expert insights to support the council’s work and ensure a collaborative approach to addressing the energy challenges and opportunities presented by AI. Among the prominent organisations joining the council are EDF, Scottish Power, National Grid, technology giants Google, Microsoft, Amazon Web Services (AWS), and chip designer ARM, as well as infrastructure investment firm Brookfield. This collaborative framework, uniting the energy and technology sectors, aims to ensure seamless coordination in speeding up the connection of energy projects to the national grid. This is particularly crucial given the increasing number of technology companies announcing plans to build data centres across the UK. Alison Kay, VP for UK and Ireland at AWS, said: “At Amazon, we’re working to meet the future energy needs of our customers, while remaining committed to powering our operations in a more sustainable way, and progressing toward our Climate Pledge commitment to become net-zero carbon by 2040. “As the world’s largest corporate purchaser of renewable energy for the fifth year in a row, we share the government’s goal to ensure the UK has sufficient access to carbon-free energy to support its AI ambitions and to help drive economic growth.” Jonathan Brearley, CEO of Ofgem, added: “AI will play an increasingly important role in transforming our energy system to be cleaner, more efficient, and more cost-effective for consumers, but only if used in a fair, secure, sustainable, and safe way. “Working alongside other members of this Council, Ofgem will ensure AI implementation puts consumer interests first – from customer service to infrastructure planning and operation – so that everyone feels the benefits of this technological innovation in energy.” This initiative aligns with the government’s Clean Power Action Plan, which focuses on connecting more homegrown clean power to the grid by building essential infrastructure and prioritising projects needed for 2030. The aim is to clear the grid connection queue, enabling crucial infrastructure projects – from housing to gigafactories and data centres – to gain access to the grid, thereby unlocking billions in investment and fostering economic growth. Furthermore, the government is streamlining planning approvals to significantly reduce the time it takes for infrastructure projects to get off the ground. This accelerated process will ensure that AI innovators can readily access cutting-edge infrastructure and the necessary power to drive forward the next wave of AI advancements. (Photo by Vlad Hilitanu) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.

Following the release of ChatGPT’s new image-generation tool, user activity has surged; millions of people have been drawn to a trend whereby uploaded images are inspired by the unique visual style of Studio Ghibli. The spike in interest contributed to record use levels for the chatbot and strained OpenAI’s infrastructure temporarily. Social media platforms were soon flooded with AI-generated images styled after work by the renowned Japanese animation studio, known for titles like Spirited Away and My Neighbor Totoro. According to Similarweb, weekly active ChatGPT users passed 150 million for the first time this year. OpenAI CEO Sam Altman said the chatbot gained one million users in a single hour in early April – matching the numbers the text-centric ChatGPT reached over five days when it first launched. SensorTower data shows the company also recorded a jump in app activity. Weekly active users, downloads, and in-app revenue all hit record levels last week, following the update to GPT-4o that enabled new image-generation features. Compared to late March, downloads rose by 11%, active users grew 5%, and revenue increased by 6%. The new tool’s popularity caused service slowdowns and intermittent outages. OpenAI acknowledged the increased load, with Altman saying that users should expect delays in feature roll-outs and occasional service disruption as capacity issues are settled. Legal questions surface around ChatGPT’s Ghibli-style AI art The viral use of Studio Ghibli-inspired AI imagery from OpenAI’s ChatGPT has raised concerns about copyright. Legal experts point out that while artistic styles themselves may not always be protected, closely mimicking a well-known look could fall into a legal grey area. “The legal landscape of AI-generated images mimicking Studio Ghibli’s distinctive style is an uncertain terrain. Copyright law has generally protected only specific expressions rather than artistic styles themselves,” said Evan Brown, partner at law firm Neal & McDevitt. Miyazaki’s past comments have also resurfaced. In 2016, the Studio Ghibli co-founder responded to early AI-generated artwork by saying, “I am utterly disgusted. I would never wish to incorporate this technology into my work at all.” OpenAI has not commented on whether the model used for its image generation was trained on content similar to Ghibli’s animation. Data privacy and personal risk The trend has also drawn attention to user privacy and data security. Christoph C. Cemper, founder of AI prompt management firm AIPRM, cautioned that uploading a photo for artistic transformation may come with more risks than many users realise. “When you upload a photo to an AI art generator, you’re giving away your biometric data (your face). Some AI tools store that data, use it to train future models, or even sell it to third parties – none of which you may be fully aware of unless you read the fine print,” Cemper said. OpenAI’s privacy policy confirms that it collects both personal information and use data, including images and content submitted by users. Unless users opt out of training data collection or request deletion via their settings, content will be retained and used to improve future AI models. Cemper said that once a facial image is uploaded, it becomes vulnerable to misuse. That data could be scraped, leaked, or used in identity theft, deepfake content, or other impersonation scams. He also pointed to prior incidents where private images were found in public AI datasets like LAION-5B, which are used to train various tools like Stable Diffusion. Copyright and licensing considerations There are also concerns that AI-generated content styled after recognisable artistic brands could cross into copyright infringement. While creating art in the style of Studio Ghibli, Disney, or Pixar might seem harmless, legal experts warn that such works may be considered derivative, especially if the mimicry is too close. In 2022, several artists filed a class-action lawsuit against AI companies, claiming their models were trained on original artwork without consent. The cases reflect the broader conversation around how to balance innovation with creators’ rights as generative AI becomes more widely used. Cemper also advised users to review carefully the terms of service on AI platforms. Many contain licensing clauses with language like “transferable rights,” “non-exclusive,” or “irrevocable licence,” which allow platforms to reproduce, modify, or distribute submitted content – even after the app is deleted. “The rollout of ChatGPT’s 4o image generator shows just how powerful AI has become as it replicates iconic artistic styles with just a few clicks. But this unprecedented capability comes with a growing risk – the lines between creativity and copyright infringement are increasingly blurred,” Cemper said. “The rapid pace of AI development also raises significant concerns about privacy and data security. There’s a pressing need for clearer, more transparent privacy policies. Users should be empowered to make informed decisions about uploading their photos or personal data.” Search interest in “ChatGPT Studio Ghibli” has increased by more than 1,200% in the past week, but alongside the creativity and virality comes a wave of serious problems about privacy, copyright, and data use. As AI image tools get more advanced and accessible, users may want to think twice before uploading personal images, especially if they’re not sure where the data may ultimately end up. (Image by YouTube Fireship) See also: Midjourney V7: Faster AI image generation Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo. Explore other upcoming enterprise technology events and webinars powered by TechForge here.