Meta CEO Mark Zuckerberg is aggressively promoting a future where AI becomes the dominant form of social interaction, claiming that AI friends, therapists, and business agents will soon outnumber human relationships. During a recent media blitz across multiple podcasts and a Stripe conference appearance, Zuckerberg cited statistics suggesting "the average American has fewer than three friends" while claiming people desire "meaningfully more, like 15 friends" -- positioning AI companions as the solution to this gap. The Meta founder's vision extends beyond casual interaction to therapeutic and commercial relationships, with personalized AI that "has a deep understanding of what's going on in this person's life." Meta has already deployed its AI across Instagram, Facebook, and Ray-Ban smart glasses, reaching nearly a billion monthly users. Read more of this story at Slashdot.

HairFlow can simulate wind, gravity changes, frizz and more.

Minding our manners is something most of us do automatically – even when speaking to AI, it seems. But saying “please” and “thank you” to AI is costing us, and our planet, as revealed by OpenAI CEO Sam Altman. Altman recently responded to a question on X asking how much money Open AI loses because of people using please and thank you in their prompts. “Tens of millions of dollars,” he replied. But it’s not just the cost that’s the issue – AI is having a direct and detrimental effect on the environment. The International Energy Agency’s electricity report revealed a request to ChatGPT consumes 10 times the electricity of a Google search, and the 2024 Google environmental report revealed its greenhouse gas emission levels had risen by 48% since 2019, directly attributing this rise to the energy demand from AI centres. Jeremy Kahn, AI editor at Fortune magazine and author of Mastering AI: A survival guide to our superpowered future, spoke to BBC Radio 4’s The Briefing Room about the threat AI poses to our planet. He said that the average data centre uses 200 megawatts in a year – approximately the same amount of energy used to power 200,000 UK homes, or 10 NHS hospitals. And that’s just one data centre. He also said that OpenAI has discussed opening a data centre which uses five gigawatts per year, or the amount of energy needed to power Miami. “It’s really adding up,” AI scientist Sasha Luccioni told the BBC’s Tech Life. “This incremental usage and switching from good old-fashioned AI to generative AI is causing a sort of ripple effect on energy grids and data centre usage. We hear a lot about AI in regard to creativity and independent thinking, ownership and whether it’s coming to take our jobs (which it is, according to Fiver CEO Micha Kaufman). These are all valid concerns. But there’s a whole other side to AI which is just as scary, and still isn’t being talked about enough in the design world. “Creatives shape the soul of society, so watching creative leaders bow down unquestioningly to AI should have us all worried.” Sustainability has become one of the design’s main concerns, thanks to the pioneering work of studios like Thomas Matthews and the Design Council, which is hosting the World Design Congress this September around its “design for planet” theme. But up until now, the focus has largely been on physical design – on materials and carbon emissions and recyclability. The Dublin-based studio Aad recently launched its Digital Sustainability Resource to highlight the importance of digital designers prioritising sustainability in their work too. It wanted to help designers understand the scale of the problem, and think more carefully about the impact their work has, as well as providing practical tips and advice. With the rise of AI, and its increasing prevalence in the digital design space, it’s important for design studios to fully and honestly consider the sustainability of their work, and push back against environmentally damaging practices. “Designers have a unique responsibility,” explains Cat Robertson, Aad’s sustainability and brand lead. “As AI tools look to democratise design and content creation, designers face a kind of dual challenge: adapting to these tools while also guiding their responsible, environmentally sustainable use.” She recommends designers push for transparency from AI developers on their environmental footprint, and establish professional standards for sustainable, ethical AI usage within design practices. “The future of design isn’t just about what we create, but how consciously we create it,” Robertson says. “Our responsibility extends beyond aesthetics to include the full lifecycle impact of our work – especially as these impacts scale through technologies like AI.” If designers care about the future of the planet, they can’t ignore what AI is doing to the environment. “Creatives shape the soul of society, so watching creative leaders bow down unquestioningly to AI should have us all worried,” says Sana Iqbal, founder of Studio Sana. “This isn’t about adapting with the times, but ensuring the AI transition isn’t exploitative and is fair to all,” she continues. “Transparency will ensure brands are meeting their regulatory requirements and not green-washing.” Designers have a responsibility to promote ethical and sustainable practices in their work – both the process, and the output. They need to be aware of how AI impacts the environment, and use it mindfully. Skipping “please” and “thank you” is a start. But we may need a more impolite conversation on what AI means for our sustainable design future.

#NintendoSwitch2 Editions and upgrade packs for #SuperMarioPartyJamboree, The Legend of #Zelda: #BreathOfTheWild, The Legend of Zelda: #TearsOfTheKingdom, and more, are also available to digitally pre-order on our website.Learn more: https://ntdo.com/6013f0QBN

IntroductionHello, my name is Valentina. I've been working as a 3D Character Artist since 2021. Before working in 3D, I studied at an art university as an art restorer. In my final university years, I learned about such a direction as 3D graphics. I was playing a lot of video games at that time, and I was quickly drawn to the prospect of creating models for video games. I studied everything to get into the game industry as a character artist. When I got a job in the industry, I became interested in expressing myself through 3D work. I was both learning new techniques to boost my skills as an artist as well as starting to come up with my own ideas for my own enjoyment.From the beginning, I have been a self-taught artist using tutorials on the Internet. However, I learned important skills while working in an outsourcing studio as a junior 3D artist with good lead artists. After that, I improved my skills again, but more through practicing making more complex models. Sometimes, but much less often than before, I also turned to various tutorials on the Internet to become a more professional artist.During this time, I tried my hand at different projects as a character artist. I've worked in gamedev, on cinematics, and for 3D printing. A famous project I was a part of was Spider-Man 2.The Shrek ProjectWhen I started work on Shrek, I thought about the fact that I wanted to do something funky. I wanted to practice my armor sculpting skills, but I also wanted to just have fun. A realistic dark fantasy-style Shrek is pretty fun, in my opinion. In this project, I tried to imagine what Shrek would look like if he were a bit more realistic, but I didn't want to lose the likeness to the original, so I tried not to change his appearance and proportions too much.ReferencesMy references included many screenshots from the original cartoon so I wouldn't forget what the original Shrek looks like. I also found references for armor of approximately what type I would like. I didn't want to copy something real, but looking at pictures of real armor, I decided that I wanted to make armor with more patterns. I also found some references of aged clothing. I checked ArtStation to see how other artists have done Shrek and thought about what I liked about their work and what I didn't like so much to decide how I wanted to stylize my work. I also just looked at a lot of different dark fantasy artworks for inspiration. In this work, I mainly wanted to try to create my interpretation of a well-known character in a different style for him: dark fantasy.ModelingThis is what the process looked like under the hood:For both character models, I started with basic meshes to speed up the creation process. I made Shrek in T-pose first because it was easier for me to control the process of creating the other elements. But Fiona could be made immediately in the required pose, as she doesn't have a lot of additional details. By placing her on Shrek's shoulder, I sewed the clothes in Marvelous.For the dress, I used two materials:Default for simulation (green material) with particle distances of 10Silk Organza (pink material) with particle distances of 5For places where I wanted to get softer and more airy pleats, I also made seams for clothes right away in Marvelous. I think it's easier and looks more convincing than adding them later in the sculpting process. The seams are added with the internal lines at different angles.All armor has a basic, simple low-poly base created in Zbrush, and then detailing is done on high subdives using a layer system. The detailing looks a bit chaotic because of the number of patterns, but I added each new type of pattern separately on each layer so that I could control the strength of the extrusion of the different types of patterns. For example, I ended up making the pattern more visible on the edges than in the center.Basically, the details were made with projections alpha, but I also added a small pattern on the edges of the helmet and a large pattern in the center of a separate object to make it more visible. On top of the alpha, I sculpted armor using Orb brushes. Specifically, Orb Flatten to create flat areas and orb slash/orb cracks for scratches. On top of that, in some places, I added a little bit of different noise. On the rest of the armor elements, I used the same method using the orb brushes. For the scratches on the dragon, I used orb cracks.In general, in this project, I got a lot of subtools; I tried to do everything as much as possible on different subtools for easy editing, collecting subtools on thematic folders. But in the end, it certainly took a long time to combine it and export.I sculpted small clothes like gloves and shoes by hand, starting with a simple shape. All folds were added with a standard brush. DamStandart did creases, damage, and torn edges. In the end, the noise was added with alphas.I edited the Marvelous garment a bit, sculpting the body to fit the pose and adding torn edges and fabric damage.To speed up the work of creating fabric damage, I created my custom brush alphas. I made the threads on the edges of the fabric using fiber mesh. Simply select a mask border of the garment and create a fiber mesh in place of the mask, adjusting the gravity down, thickness, amount, and length. The extra threads between the fabric I added using a curved tube.I made the skin using Super Skin – 41 ZBrush Brushes Set for Realistic Human Skin.TopologyThis is not a game-ready model, so I didn't do hand retopology for this model. For subtools with subdivisions like clothes, armor, and body, I just exported the first subdivision and automatically unwrapped it in RizomUV. Some subtools I lightened with decimation and also did an automatic UV.TexturingThe main task of texturing Shrek's clothes was to make them dirty and greasy, with a lot of stains, different from Princess Fiona's dress, which is absolutely clean because she was just stolen from the castle. I made the dirt in different colors so it wouldn't look the same.Some of it is brown, some of it is reddish, some of it is yellowish, and some of it is blue spots, which is good to dilute the overall color palette of the shirt. With the white color, I lightened the scuff marks a bit. Shrek's skin was made of material from Substance Skin Face. Next, using an empty layer in passthrough mode with levels base color and HSL Perceptive filter, the color was changed from flesh to green. Next comes a layer with a translucent flesh texture to diversify the homogeneous skin tone.Then, using the ambient occlusion and light generators, I darkened the depressions on the face. I added different spots and dirt on the face with the spot brush manually. I also added some color variations manually with a soft brush. Again, I did a little color correction on all of the existing layers. With a curvature filter with a dark color, I emphasized sculpted pores and small hollows on the face. At the end I added some rougness to Shrek with fill layer dirt4. I decided it suited Shrek to look a little sweaty.For armor, just like with clothes and leather, I took a base of some ready-made material and then added color variations according to the same principle. I wanted to keep some balance between still shiny, beautiful armor and already looking a bit old lying down. It's not new armor, after all; Shrek picked up someone's armor that had been lying around for a long time. In the shoulder armor and helmet, I wanted to keep a reference to the dragon they were escaping from to further tie those events together in this scene.Lighting & RenderingI used four light sources. Directional for the main bright light and omni for additional illumination of the model from the top and back/side of the model.In the camera settings, I edited curve RGB to edit the color tone a bit and make the image contrast on the curve. Next, I added the fog. I added sharpen to make the details stand out better. And, of course, I added a vignette. I prefer a 50 mm perspective. I also used sticky focus for focus and added a little chromatic aberration.In post-production, I tried not to change the image too much, but I did make a couple of edits. I made the image a cooler tone. I added darkening at the bottom using a gradient on the soft light layer. With the lightener tool, I lightened the top of the model a bit. And at the end, I added some gaussian noise.ConclusionIt's hard to say exactly how long this project took because I took quite a few breaks, including a break for another project. In total, with these breaks, I worked for 4 months.The main difficulty is probably the lack of any supporting concept, so I had to keep all the details in my head, and in the process, I invented and implemented new ones. For example, from the very beginning I thought to make just Shrek in armor without Fiona, but now having the final image, I can't imagine this model without Fiona on his shoulder.I learned to be more flexible in coming up with my own ideas in the process. And quite patient, this is probably one of my longest personal projects. I've also learned how to interpret well-known characters in a different style while keeping them recognizable. And got better at sculpting armor. I advise beginners to do what they like first, think about how to apply their art and learn the necessary skills for further work.Thank you for reading, bye!Valentina Iudina, 3D Character & Creature ArtistInterview conducted by Emma Collins

Eric "ConcernedApe" Barone prefers just adding stuff to Stardew Valley via updates, but the game's creator isn't ruling out Stardew Valley 2 in the future. However, that could be a ways off, since he's hoping his next game, Haunted Chocolatier, will arrive in the next five years.Appearing on Bobby Lee's TigerBelly podcast (spotted by GamesRadar), Barone was asked why he doesn't just make Stardew Valley 2. The Stardew Valley creator explained why updates are more appealing to him than starting entirely over. "It's so much easier to just add more stuff to Stardew Valley than to make a whole new game from scratch," said Barone. "Stardew Valley, all of the major systems are already done." But Barone did give hope to people wanting a full-fledged sequel in the future. "I might eventually make a Stardew Valley 2, to be honest," Barone said. First, though, he wants to switch things up. "I don't want to just be the Stardew Valley guy, that's why I'm making Haunted Chocolatier," he said.Continue Reading at GameSpot

Sirene is one of two Axons that players must defeat in Clair Obscur: Expedition 33, and they will have the opportunity to collect a variety of items while pursuing this powerful foe. Among those items are weapons, Pictos, and upgrade materials, and some players will undoubtedly want to find them all. This guide is written with those players in mind, and it will help them collect every item in Clair Obscur: Expedition 33's Sirene.

As AI use cases continue to expand — from document summarization to custom software agents — developers and enthusiasts are seeking faster, more flexible ways to run large language models (LLMs). Running models locally on PCs with NVIDIA GeForce RTX GPUs enables high-performance inference, enhanced data privacy and full control over AI deployment and integration. Tools like LM Studio — free to try — make this possible, giving users an easy way to explore and build with LLMs on their own hardware. LM Studio has become one of the most widely adopted tools for local LLM inference. Built on the high-performance llama.cpp runtime, the app allows models to run entirely offline and can also serve as OpenAI-compatible application programming interface (API) endpoints for integration into custom workflows. The release of LM Studio 0.3.15 brings improved performance for RTX GPUs thanks to CUDA 12.8, significantly improving model load and response times. The update also introduces new developer-focused features, including enhanced tool use via the “tool_choice” parameter and a redesigned system prompt editor. The latest improvements to LM Studio improve its performance and usability — delivering the highest throughput yet on RTX AI PCs. This means faster responses, snappier interactions and better tools for building and integrating AI locally.  Where Everyday Apps Meet AI Acceleration LM Studio is built for flexibility — suited for both casual experimentation or full integration into custom workflows. Users can interact with models through a desktop chat interface or enable developer mode to serve OpenAI-compatible API endpoints. This makes it easy to connect local LLMs to workflows in apps like VS Code or bespoke desktop agents. For example, LM Studio can be integrated with Obsidian, a popular markdown-based knowledge management app. Using community-developed plug-ins like Text Generator and Smart Connections, users can generate content, summarize research and query their own notes — all powered by local LLMs running through LM Studio. These plug-ins connect directly to LM Studio’s local server, enabling fast, private AI interactions without relying on the cloud. Example of using LM Studio to generate notes accelerated by RTX. The 0.3.15 update adds new developer capabilities, including more granular control over tool use via the “tool_choice” parameter and an upgraded system prompt editor for handling longer or more complex prompts. The tool_choice parameter lets developers control how models engage with external tools — whether by forcing a tool call, disabling it entirely or allowing the model to decide dynamically. This added flexibility is especially valuable for building structured interactions, retrieval-augmented generation (RAG) workflows or agent pipelines. Together, these updates enhance both experimentation and production use cases for developers building with LLMs. LM Studio supports a broad range of open models — including Gemma, Llama 3, Mistral and Orca — and a variety of quantization formats, from 4-bit to full precision. Common use cases span RAG, multi-turn chat with long context windows, document-based Q&A and local agent pipelines. And by using local inference servers powered by the NVIDIA RTX-accelerated llama.cpp software library, users on RTX AI PCs can integrate local LLMs with ease. Whether optimizing for efficiency on a compact RTX-powered system or maximizing throughput on a high-performance desktop, LM Studio delivers full control, speed and privacy — all on RTX. Experience Maximum Throughput on RTX GPUs At the core of LM Studio’s acceleration is llama.cpp — an open-source runtime designed for efficient inference on consumer hardware. NVIDIA partnered with the LM Studio and llama.cpp communities to integrate several enhancements to maximize RTX GPU performance. Key optimizations include: CUDA graph enablement: Groups multiple GPU operations into a single CPU call, reducing CPU overhead and improving model throughput by up to 35%. Flash attention CUDA kernels: Boosts throughput by up to 15% by improving how LLMs process attention — a critical operation in transformer models. This optimization enables longer context windows without increasing memory or compute requirements. Support for the latest RTX architectures: LM Studio’s update to CUDA 12.8 ensures compatibility with the full range of RTX AI PCs — from GeForce RTX 20 Series to NVIDIA Blackwell-class GPUs, giving users the flexibility to scale their local AI workflows from laptops to high-end desktops. Data measured using different versions of LM Studio and CUDA backends on GeForce RTX 5080 on DeepSeek-R1-Distill-Llama-8B model. All configurations measured using Q4_K_M GGUF (Int4) quantization at BS=1, ISL=4000, OSL=200, with Flash Attention ON. Graph showcases ~27% speedup with the latest version of LM Studio due to NVIDIA contributions to the llama.cpp inference backend. With a compatible driver, LM Studio automatically upgrades to the CUDA 12.8 runtime, enabling significantly faster model load times and higher overall performance. These enhancements deliver smoother inference and faster response times across the full range of RTX AI PCs — from thin, light laptops to high-performance desktops and workstations. Get Started With LM Studio LM Studio is free to download and runs on Windows, macOS and Linux. With the latest 0.3.15 release and ongoing optimizations, users can expect continued improvements in performance, customization and usability — making local AI faster, more flexible and more accessible. Users can load a model through the desktop chat interface or enable developer mode to expose an OpenAI-compatible API. To quickly get started, download the latest version of LM Studio and open up the application. Click the magnifying glass icon on the left panel to open up the Discover menu. Select the Runtime settings on the left panel and search for the CUDA 12 llama.cpp (Windows) runtime in the availability list. Select the button to Download and Install. After the installation completes, configure LM Studio to use this runtime by default by selecting CUDA 12 llama.cpp (Windows) in the Default Selections dropdown. For the final steps in optimizing CUDA execution, load a model in LM Studio and enter the Settings menu by clicking the gear icon to the left of the loaded model. From the resulting dropdown menu, toggle “Flash Attention” to be on and offload all model layers onto the GPU by dragging the “GPU Offload” slider to the right. Once these features are enabled and configured, running NVIDIA GPU inference on a local setup is good to go. LM Studio supports model presets, a range of quantization formats and developer controls like tool_choice for fine-tuned inference. For those looking to contribute, the llama.cpp GitHub repository is actively maintained and continues to evolve with community- and NVIDIA-driven performance enhancements. Each week, the RTX AI Garage blog series features community-driven AI innovations and content for those looking to learn more about NVIDIA NIM microservices and AI Blueprints, as well as building AI agents, creative workflows, digital humans, productivity apps and more on AI PCs and workstations. Plug in to NVIDIA AI PC on Facebook, Instagram, TikTok and X — and stay informed by subscribing to the RTX AI PC newsletter. Follow NVIDIA Workstation on LinkedIn and X.

Attention, heroic Helldivers.  To celebrate the eradication of the Illuminate invaders, Super Earth mobilizes our most regal looking troops yet – a true celebration of our forebears. Enter the Masters of Ceremony Premium Warbond*! Wield all-new elite weaponry, capes and armours to step into the boots of Super Earth’s most decorated units.  Play Video To start, let us look at the R-2 Amendment. Carve through bugs like butter with an attached bayonet for close-quarter combat alongside a 20-round magazine. Perfect for vanquishing foes in style both up close and at a distance. Paired with the CQC-2 Sabre. A sabre hailing from an elegant time will inspire you to lead the charge and thrust your blade through the heart of our enemies.  To round out the parade spectacle; introducing the G-142 Pyrotech Grenade.  When thrown, it will spin and shower the surrounding area in sparks before ceremoniously exploding in a ball of flames. Farewell, pitiful enemies of Super Earth. Show off your love for Managed Democracy and inspire those around you with the CQC-1 One True Flag.  Feared across the galaxy and attesting to the might of Managed Democracy, this sharp symbol of freedom doubles as a spear to drive home the values of our glorious sovereignty. It brings a tear to our eyes. Sure, your killing looks great. But how do you look, Helldiver? Let’s accessorize.  For formal occasions the RE-1861 Parade Commander brings old-school prestige whilst being lightweight enough to react in the heat of battle. Pair it with the Humble Regalia cape to make it shine for Democracy in red, black and gold. After something a little more menacing? Look no further than the RE-2310 Honorary Guard. A traditional but battle hardened armor. Pair with the Federations Embrace cape to pay honor to the glory of Super Earth and our fallen forebears.  Both armors come stock standard with the Reinforced Epaulettes perk, offering your limbs the chance of not breaking so you can continue to wave the flag of victory. For maximum respect, unlock the Decorated Hero title and Lockstep emote to ensure you and your fellow Helldivers parade together on the bodies of your victories. In unison. An intricate new look from Warbonds of past eras By popular demand, we are adding a new stash of FRV patterns to previously released Warbonds to all owners. Go check out the Viper Commando, Freedom’s Flame, Chemical Agents, and Truth Enforcers Warbond pages in Acquisitions on May 15! Are you ready to join the rank of heroes? The Masters of Ceremony and new FRV patterns march in on May 15 for PS5 and PC.

2K Games and developer Hangar 13 revealed a longer look at gameplay from Mafia: The Old Country on Thursday, showing how the origin story for the Mafia franchise stands apart from the open-world, organized crime competition. In fact, Hangar 13 is hinting strongly that it has no intentions to make Mafia: The Old Country your next “forever game.” Instead, the game’s creators call The Old Country “a focused, linear experience that combines quality storytelling, authentic era immersion and a refined take on the familiar Mafia gameplay.” Hangar 13 president Nick Baynes said in a news release that the smaller, more linear approach “allows us to deliver a story that’s gritty, grounded, brutal and emotional.” Further reinforcing the smaller, less-bloated design of Mafia: The Old Country is this statement from the president of publisher 2K, David Ismalier, who said, “We think there’s a large audience for compelling stories that don’t require massive time commitments.” Ismailer added that The Old Country “can easily complement the other more persistent games our players also love and engage with on a more consistent basis.” The strongest evidence of Mafia: The Old Country’s scope is its price: 2K is selling the base game for $49.99. There’s a Deluxe Edition that costs $59.99 and includes a bunch of extras, naturally, but the asking price is certainly atypical of AAA game releases in 2025. So if you’re in for a leaner, more classic Mafia experience, check out the newly released gameplay video that shows protagonist Enzo Favara, as he takes the oath, works his way up Don Torrisi’s crime family, and shoots up 1900s Sicily. Mafia: The Old Country is coming to PlayStation 5, Windows PC, and Xbox Series X on Aug. 8.