Volkswagen teased this concept image of its forthcoming entry-level EV.Volkswagen has announced that next month, it will show off a new low-cost, entry-level EV that could cost just 20,000 (about $20,800 USD), the price it suggested it would hit back when it announced the ID. Life concept in 2021. Thats a steal compared to current low-end US EVs like the over-$28,000 Nissan Leaf. Meanwhile, Volkswagens most affordable American-sold electric car is the roughly $40,000 ID.4, despite having promised a more affordable car for so long.The company didnt say what the cheaper follow-up model will be called, but adjusting the concept image VW included with its announcement exposes the name ID.One, as one Bluesky user found. That tracks: Volkswagen previously namedropped an ID.1 that would be based on the ID. Life concept. In 2022, VW shared a sketch of that car and pegged it for a 2027 release. This new car will hit in 2027, says VW, so theyre likely one and the same.rumor has it ID.ONE sir Felix Hamer electricfelix (@electricfelix.bsky.social) 2025-02-05T22:00:02.499ZLow-cost entry-level mobility in the electric era will be one of the cornerstones of the brands future plan, Volkswagen writes, adding that its not planning to just produce one low-cost car it reiterated today that it will ship another cheap EV called the ID. 2all that it announced almost two years ago. That car is expected to go for 25,000 (or about $26,000 USD). Both it and the ID.1 (or ID.One, if you like) will use a new version of its modular battery platform that the ID.4 and ID Buzz are based on.EV prices arent the only place Volkswagen is trying to cut costs. Following a labor strike in Germany, it reached a union deal that will mean 35,000 fewer workers and billions per year in cost-cutting, The Wall Street Journal notes.

OpenAIs Deep Research AI Agent offers a powerful research assistant at a premium price of $200 per month. However, the open-source community has stepped up to provide cost-effective and customizable alternatives. Here are four fully open-source AI research agents that can rival OpenAIs offering:1. Deep-ResearchOverview:Deep-Research is an iterative research agent that autonomously generates search queries, scrapes websites, and processes information using AI reasoning models. It aims to provide a structured approach to deep research tasks.Key Features:Query Generation: Dynamically generates optimized search queries.Web Scraping with Firecrawl: Extracts useful information from websites.o3-Mini Model for Reasoning: Uses OpenAIs o3-mini model for intelligent processing.100% Open Source: Fully accessible and modifiable.GitHub Repository: https://github.com/dzhng/deep-research2. OpenDeepResearcherOverview:OpenDeepResearcher is an asynchronous AI research agent designed to conduct comprehensive research iteratively. It utilizes multiple search engines, content extraction tools, and LLM APIs to provide detailed insights.Key Features:SERP API Integration: Automates iterative search queries.Jina AI for Content Extraction: Extracts and summarizes webpage content.OpenRouter LLM Processing: Utilizes various open LLMs for reasoning.100% Open Source: Offers flexibility in customization and deployment.GitHub Repository: https://github.com/mshumer/OpenDeepResearcher3. Open Deep Research by FirecrawlOverview:Open Deep Research is a lightweight and efficient AI research agent that leverages Firecrawl search and extraction mechanisms. It allows users to reason with any LLM of their choice rather than relying on a fine-tuned proprietary model.Key Features:Firecrawl Search + Extract: Fetches and extracts relevant content efficiently.Customizable AI Reasoning: Supports any LLM via the AI SDK.Open Source & Self-Hostable: Full control over deployment and customization.GitHub Repository: https://github.com/nickscamara/open-deep-research4. DeepResearch by Jina AIOverview:DeepResearch by Jina AI is an advanced AI research assistant that replicates OpenAIs agentic search, read, and reasoning workflow. It integrates multiple search engines and employs an AI-driven approach to extract and summarize relevant information.Key Features:Search Integration: Uses Gemini Flash, Brave, and DuckDuckGo for diverse search results.AI-Powered Reading: Implements Jina Reader to extract and summarize content efficiently.Reasoning Process: Uses advanced AI models for contextual understanding.100% Open Source: Fully customizable and self-hostable.GitHub Repository: https://github.com/jina-ai/node-DeepResearchConclusionThese four open-source AI research agents provide powerful alternatives to OpenAIs Deep Research AI Agent. With robust search capabilities, AI-powered extraction, and reasoning features, they enable researchers to automate and optimize their workflows without incurring high costs. Since all options are open-source, users have complete flexibility to modify, extend, and self-host these tools based on their specific needs.This article is inspired from this Tweet. Also,dont forget to follow us onTwitterand join ourTelegram ChannelandLinkedIn Group. Dont Forget to join our75k+ ML SubReddit. Aswin AkAswin AK is a consulting intern at MarkTechPost. He is pursuing his Dual Degree at the Indian Institute of Technology, Kharagpur. He is passionate about data science and machine learning, bringing a strong academic background and hands-on experience in solving real-life cross-domain challenges.Aswin Akhttps://www.marktechpost.com/author/aswinak/Meet Satori: A New AI Framework for Advancing LLM Reasoning through Deep Thinking without a Strong Teacher ModelAswin Akhttps://www.marktechpost.com/author/aswinak/Meta AI Introduces VideoJAM: A Novel AI Framework that Enhances Motion Coherence in AI-Generated VideosAswin Akhttps://www.marktechpost.com/author/aswinak/Google DeepMind Researchers Unlock the Potential of Decoding-Based Regression for Tabular and Density Estimation TasksAswin Akhttps://www.marktechpost.com/author/aswinak/Top AI Coding Agents in 2025 [Recommended] Join Our Telegram Channel

Semantic Search Engine Using Langchain 0 like February 5, 2025Share this postAuthor(s): Lo Zarantonello Originally published on Towards AI. Load and query a PDF locally using LangchainThis member-only story is on us. Upgrade to access all of Medium.In this post, I am loosely following Build a semantic search engine on Lnagchain, adding some explanation about Embeddings and Vector Store.We start by installing @langchain/community and pdf-parse in a new directory.npm i @langchain/community pdf-parseThe @langchain/community package contains a range of third-party integrationsThe pdf-parse package is a pure javascript cross-platform module to extract texts from PDFs.Below, you can see how @langchain/community fits into the Langchain ecosystem.Langchain ecosystemIn the same directory, we can create a new index.js file and add the following code.import { PDFLoader } from "@langchain/community/document_loaders/fs/pdf";const loader = new PDFLoader("./pdfs/letter-to-shareholders-amazon.pdf");// loads one Document object per PDF pageconst docs = await loader.load();console.log(docs.length);PDFLoader loads one Document object per PDF page. So, in my case, docs is an array of 8 Document objects because the PDF is 8 pages long.// Document object{ pageContent: String, metadata: { source: './pdfs/letter-to-shareholders-amazon.pdf', pdf: { version: '1.10.100', info: [Object], metadata: null, totalPages: 8 }, loc: { pageNumber: 1 } }, id: undefined}To run the loader in a node environment, we can simply runnode index.js Node doesnt recognize ES modules by default so we need to add the following type field in package.json.{ "dependencies": { "@langchain/community": "^0.3.28", Read the full blog for free on Medium.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AITowards AI - Medium Share this post

NN#1 Neural Networks Decoded: Concepts Over Code 0 like February 5, 2025Share this postAuthor(s): RSD Studio.ai Originally published on Towards AI. This member-only story is on us. Upgrade to access all of Medium.Source: Analytics VidhyaWe stand at the cusp of an era where Artificial Intelligence is rapidly transforming our world. From self-driving cars navigating complex streets to AI assistants anticipating our needs, the impact of intelligent machines is becoming increasingly profound. But where did this journey begin? What were the initial sparks of inspiration that led to the creation of these sophisticated systems we call Neural Networks?This article, the first in our series Neural Networks Decoded: Concepts Over Code, takes a step back to explore the very conception of neural networks. We wont dive into code or complex equations just yet. Instead, well embark on a journey to understand the origins of this revolutionary technology, tracing its roots back to the remarkable organ that inspired it all: the human brain.Think about it: for centuries, humans have been fascinated by the mystery of their own minds. How does this intricate network of biological tissue give rise to consciousness, thought, and learning? Its this very question that sparked the initial dreams of creating artificial intelligence, and neural networks were born from the desire to mimic, in a computational form, the brains fundamental architecture.Welcome Read the full blog for free on Medium.Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming asponsor. Published via Towards AITowards AI - Medium Share this post

Speaking of Jason Snell, he and I played a round of Clockwise with Rosemary Orchard and Dan Moren today. Check out the podcast episode below. more

iPhone 7 and iPhone 7 Plus were fantastic smartphones in 2016 and 2017, but they did experience a so-called Loop Disease in some instances. For that reason, a class action lawsuit was filed years ago and eventually approved last year. A year later, former iPhone 7 customers are starting to receive their settlement payments.The amounts appear to bearound$200 per claim filed. Thats not a badpaydayfor former iPhone 7 users in 2025. Of course, those payments come from a $35 million settlement that best rewards the attorneys who brought the class action suit forward.The deadline to participate in the class action suit ended last summer. For its part, Apple admitted no wrongdoing and denied the allegations while agreeing to the settlement.Loop Disease was allegedly caused by putting pressure on a certain part of the iPhone 7 or iPhone 7 Plus. The symptoms included degraded audio quality, including during FaceTime calls, and more.Actual payments receiveddependon whether a participant paid for a repair related to the issue. Customers who experienced the issue but didnt pay for a repair are expected to receive less. The top payment amount was said to be $350 at the time.Last fall, customers of MacBooks with butterfly keyboards similarly began receiving payment as part of a separate settlement.Receive your settlement payment? Still rock an iPhone 7? Experiencing a different problem with your current iPhone? Let us know in the comments below.Top iPhone accessoriesAdd 9to5Mac to your Google News feed. FTC: We use income earning auto affiliate links. More.Youre reading 9to5Mac experts who break news about Apple and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Mac on Twitter, Facebook, and LinkedIn to stay in the loop. Dont know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

Billionaire Elon Musk's overthrow of the US Treasury a move many insiders believe to be illegal by his so-called Department of Government Efficiency has even some of president Donald Trump's loyalists uneasy.The ultra-public rampage, Wired reports, is sowing discord between Republican aides and Musk's army of strikingly young men, some still in college, who are currently making astonishingly deep changes to the federal government's payments system without clearance."I think its more the staff who have an issue with Elon than President Trump," one Republican aide told the publication."There could be a collision course coming here at some point," the staffer added, pointing out that some believe Musk is "getting too big for his breeches."According to Wired's reporting, there's growing concern that the blinding number of policy decisions are being made at such a breakneck pace that even the president's office isn't capable of keeping track an unease that highlights just how chaotic and destructive the first two weeks of Trump's tenure have been.Many have also pointed out that Musk and Trump's on-and-off-again relationship could be greatly strained by having the former actively pull the purse strings and assume control actions that could directly undermine the agency of the president himself.Trump also appears to be fairly clueless about what it is DOGE is actually doing. The president told Fox News on Tuesday that he had no idea where DOGE staffers were working, despite getting office space inside the White House itself, as flagged by Wired.Nonetheless, a source close to Trump told the publication that the president is "entirely enthusiastic about [Musk's] efforts, and they are working together very closely."Intriguingly, Trump directly addressed the subject of playing second fiddle to Musk, telling reporters that the billionaire "can't do and won't do" anything "without our approval."Where all of this drama leaves the role of Trumpworld figures who have been subjugated to watching the chaos unfurl remains to be seen. The GOP Congress appears to have little interest in stepping in to restrain Musk, as Semafor reports, suggesting they're too afraid to get caught up in the crossfire.However, some lawmakers argue Republicans are still paying attention."I can see things where Musk hes coming up with good ideas he could go too far," North Carolina Republican senator Thom Tillis told Semafor. "We could say, 'Great idea. It doesnt work in a public institution.'"But given DOGE's indiscriminate dismantling of institutions including USAID, Republicans clearly have yet to reach such a point.Senator Jeanne Shaheen (D-NH) told Semafor that Republicans are "afraid to say" that undercutting USAID is a terrible idea, saying they're"afraid of retribution."If there's one certainty, it's that the playbook has long been thrown out along, according to some observers, with certain strictures of the US Constitution.More on DOGE: Elon Musk Says DOGE Will Now Shut Down Government Payments He Doesn't LikeShare This Article

Software Design Engineer(Remote, Full-Time, Anywhere in the World)Starting Pay: USD $50-100K/yrExact offer will be formulated based on your demonstrated skills and abilities at the time of hiring process evaluation, taking into account both development and design skills. Further pay promotions are available based on performance once on the job, with exceptionally good performance enabling promotion substantially beyond the above range.!! EXPERT DEVELOPERS ONLY !! TYPESCRIPT AND REACT EXPERIENCE REQUIRED !!THIS IS A TECHNICAL SOFTWARE DESIGN ROLE, NOT A UI / UX / PRODUCT DESIGNER ROLE !!IntroductionElite Software Automation (ESA) is a company that makes custom software solutions for fast growing enterprises with detailed business processes. The purpose of these solutions is to run intricate and highly detailed processes of our clients, and boost their business process efficiency with error minimization, process automation, and other business process enhancements.These solutions are composed of internal systems (such as CRM, ERP or other systems heavily customized and extended by us), integrated with the external systems (such as client portals) and various custom backend connections that allow interaction with external parties as needed. Our tech stack is largely based on React and Typescript (the knowledge of which is required for this role as well).Our solutions are completely custom to each client and have a lot of custom functionality in them. To deliver such solutions with high quality and high efficiency, our organization has developed very specific processes for how we run our development operation and very specific back-end and front-end toolkits to reuse functionalities and utilities and to minimize unnecessary development work for each client.The Role of Software Design Engineer is a technical solution design role within ESA. This role is reserved only for very technically skilled individuals with software development background and the actual ability to build quality working software, and an established familiarity in ESA's core tech stack based around React and Typescript. Having developed software design skills and experience will be a major plus for any candidate coming in for this role, but is secondary to technical software development abilities. Our methodology relies on skilled developers with real software building experience to design our complex and functional software solutions.Software Design Engineers on ESA team work together with the business process experts who make decisions on how the business processes of our clients need to be re-organized, re-engineered, digitized, and automated. The business process experts are responsible for the business operations and business decisions side of the matter, while the Software Design Engineers are responsible for the software functionality, design, and assuring technical implementation of a software solution that will effectively address business process requirements of the business process experts. The Software Design Engineers concept, architect, detail out, wireframe, and oversee the full functional and technical design of the software solutions. Much of the technical implementation is handled by our software development team composed of front-end and back-end software engineers, as well as QA engineers, but Software Design Engineers assure successful implementations and iterate them as needed to serve the needs of successful operation of these software solutions in our clients' businesses.Click "Apply" To Go To Our SiteThis job board has a char limit, and this job description is abridged on this site. Please click "Apply" button to go to our website and see the full job description in detail, and apply.Tech UsedWe make very intricate custom solutions and service a large number of enterprises. Over the years we have developed a very tailored tech stack to enable us to efficiently create and advance ready solutions that are both highly custom to each client's business while also being very featureful and efficient to design, develop, maintain, and iterate:- Custom Front End Development (custom portals, custom internal systems, custom interfaces, custom back-end automations, etc)React with Material UI and our custom components (Typescript is used heavily)Graphql based backend servicesREST based backend servicesOur reusable toolkits (see below)- Reusable toolkits for our operation (common reusable functionalities, common generic business functionalities, common integrations, self-developed toolkits to speed up development and QA process, etc)Custom back end and front-end toolkits developed in reusable format that we deploy across projects for reusable functionalitiesCustomized database toolkit forked on top of open source NocoDB with enhanced automatically generated endpoints and enhanced toolkits for developing and QAing any interactions with the database.Customized back-end flow toolkit for internal use forked on top of fair code N8N with enhanced endpoints toolkit accessible to front-end - very useful for developing and QA'ing background workflows efficiently.Customized scheduling toolkit forked from Cal.comOther internally developed toolkits- Third Party Systems we tightly integrate with:Zoho Platform (CRM, Desk, Books), heavily customized for each projectthese solutions are often part of our central systems solutions, and their platforms allow to use some of their out of the box functionalities, while being able to extensively customize the UI and embed custom components and interfaces into them, they are extremely customizable while providing some useful business features, which is why we use them;Customer.ioQuickbooksTwilioSendgridStripe, Authorize.netAnvil document automation toolkitOystehr healthcare backend servicesOthers, as needed to meet the needs of particular client industries, and any other needs we may have.Click "Apply" To Go To Our SiteThis job board has a char limit, and this job description is abridged on this site. Please click "Apply" button to go to our website and see the full job description in detail, and apply.Who is suitable for this Software Design Engineer role?The main trait required for this job is the ability to build something that works and successfully meets the specific business requirements. Our typical Software Design Engineer is either someone who is already doing a similar job right now (and maybe not being credited for it) or someone who has been working as a software developer but who wants to go beyond that and really make a difference in how software is designed, instead of just implementing what was designed by someone else. If you are a developer with the ability to grasp the bigger picture, but in your current job you frequently feel frustrated by bad or missing requirements, by designs with low quality standards, or that fail to consider all usage scenarios, and otherwise not being optimal, then you might be the right person for this job.The Software Design Engineer is:A builder, with intricate commercial experience of software development, understanding how it works, common software development concepts such as REST APIs, typecasting, synchronous and asynchronous execution, cache invalidation, etc., and can actually build a working software solution from the ground up.A critical thinker, who uses logic and rational thought in the analysis of business requirements, understanding the big picture as well as the finer details, and who relies on data to make decisions, asking questions when needed, instead of resorting to guesswork.A problem solver, who can split big problems into smaller, manageable chunks, and who doesn't get stuck on seemingly unsolvable problems, always finding practical, simple, feasible, effective and efficient solutions.Someone with a knack for software design, who can conceptualize functionality to meet business goals and communicate how this can be accomplished by designing user flows, low-fidelity wireframes, flow charts, etc., as well as specifying all the technical requirements for the implementation of the solution.The Software Design Engineer is:NOT a business analyst: wont write business requirements.NOT a "solutions architect": wont design vague solution architectures, instead will design highly detailed and intricate solutions, detailing all intricate front-end and back-end functionality.NOT a salesman: won't sell software nor have direct contact with customers.NOT a UX Researcher: wont interview stakeholders nor run user research.NOT a UX/UI designer: knowledge of basic UX and UI principles is required, but the job is much more than just making wireframes or prototypes.As a Software Design Engineer, what will you be doing exactly? Perform critical analysis of business requirements and business processes provided by our team of business process experts.Develop a thorough understanding of the problem and scope of the solution.Strategize the time and effort required for the implementation of the software solution and use that to construct designs that optimize for faster build-out while also meeting its requirements most effectively.Conceptualize a feasible, effective and efficient solution, balancing business goals, user goals, cost, time, and any other constraints.Perform a rigorous technical analysis of how a technically efficient and feasible solution should be built, which includes things like Database analysis, APIs analysis, external systems testing, proof of concepts construction and testing, validation of feasibility, and other intricately technicalDesign and document all aspects of the solution:Design and document the general architecture of the solution as a whole, conceptualizing its technical components and various integrations.Design and document User Flows for all user-facing functionality.Design all back-end flows and document them in flowcharts and other representations.Define all the technical specifications required to implement all functionalities and automations.Design low-fi wireframes for all user interfaces (your low-fi wireframes should be good enough to be used by developers in implementing the solution, in very rare cases high fidelity wireframes may need to be made but we almost always rely on low-fi wireframes in all of our operations).Create state diagrams, entity-relationship diagrams, flow charts, and any other technical specifications that might be required to communicate the technical requirements of your solution properly.Collate all the designs, specifications, requirements, and other documentation and hand them over to the Development Team for implementation.Assist the Development Team and the QA Team in developing, testing, and deployment.Actively participate in the implementation and assure the end goal of deploying a production-grade, fully functional, and efficient solution that meets all of the specified business requirements.Click "Apply" To Go To Our SiteThis job board has a char limit, and this job description is abridged on this site. Please click "Apply" button to go to our website and see the full job description in detail, and apply.Requirements for Candidates for this RolePractical and commercially successful Software Development experience in building working and successfully implemented software that is in active use in real world business operations;Professional knowledge and successful experience with front-end development, back-end and full-stack development, databases, APIs, and all other key software development concepts;Specific and strong experience with and knowledge of React and Typescript - this is required due to Elite Software Automation's tech stack, in this job you will sometimes find yourself working as a developer in our development team, and you must be fully capable of what our developers are capable of with our stack;Understanding of what this job is and having genuine desire and interest in utilizing your skills and experience in a Software Design Engineer capacity is required;Experience with either designing software from the ground up or with developing complex software effectively while not being given a detailed design that somebody else made (in effect coming up with design of your software while building it and working from only vague business-side-only requirements) in your past career is also required.Ability and willingness to work extremely hard. This is a very difficult job. It is also likely that it will require a lot of fast on-the-job learning for you to become good at it. If you get this job, it might be one of the hardest jobs in your life, you must be prepared for that.Click "Apply" To Go To Our SiteThis job board has a char limit, and this job description is abridged on this site. Please click "Apply" button to go to our website and see the full job description in detail, and apply.

About the opportunity:Outlier is looking for talented coders to help train generative artificial intelligence modelsThis freelance opportunity is remote and hours are flexible, so you can work whenever is best for youYou may contribute your expertise byCrafting and answering questions related to computer science in order to help train AI modelsEvaluating and ranking code generated by AI modelsExamples of desirable expertise:Currently enrolled in or completed a bachelor's degree or higher in computer science at a selective institutionProficiency working with one or more of the the following languages: Java, Python, JavaScript / TypeScript, C++, Swift, and VerilogAbility to articulate complex concepts fluently in DutchExcellent attention to detail, including grammar, punctuation, and style guidelinesPayment:Currently, pay rates for core project work by coding experts range from USD $25 to $50 per hour.Rates vary based on expertise, skills assessment, location, project need, and other factors. For example, higher rates may be offered to PhDs. For non-core work, such as during initial project onboarding or project overtime phases, lower rates may apply. Certain projects offer incentive payments. Please review the payment terms for each project.Apply NowLet's start your dream job Apply now

Google delivered big artificial intelligence news on Wednesday, launching its next-generation chatbot, Gemini 2.0. Google is opening up new Gemini 2.0 models in a multi-pronged change.Let's break down what's new and coming from Google.Gemini 2.0 Pro Upgrade your inbox Get cnet insider From talking fridges to iPhones, our experts are here to help make the world a little less complicated. The company is releasing an experimental version of Gemini 2.0 Pro inGoogle AI Studio,Vertex AIand in theGemini appfor Gemini Advanced users. The company calls this "our best model yet for coding performance and complex prompts."Gemini 2.0 FlashThe updated Gemini 2.0 Flash is now generally available via the Gemini API inGoogle AI StudioandVertex AI, meaning developers can now build production applications with 2.0 Flash. This gives more people access to the smaller, more efficient version of Google Gemini in the wake ofDeepSeek's major splash.Read more: What Is DeepSeek? Everything to Know About the New Chinese AI ToolGemini 2.0 Flash-LiteGoogle is also countering Chinese AI startupDeepSeek with Gemini 2.0 Flash-Lite, a less expensive alternative that is now available in Google AI Studio and Vertex AI for developers. (By comparison, all versions of DeepSeek's AI offerings are free.)Google says Flash-Lite (get it?) is its most cost-efficient model yet.2.0 Pro and Flash Thinking ExperimentalPeople with the Gemini app will also get to try out Google Gemini 2.0 Flash Thinking Experimental, a benchmark darling as the top AI model according toChatbot Arena's leaderboards. This version of Gemini 2.0 Flash is special, thanks to its ability to break down user prompts into a series of steps to help provide better responses to complex prompts. For those looking for something more powerful, Google also added Gemini 2.0 Pro Experimental to the Gemini app for Gemini Advanced users, as well as Google AI Studio and Vertex AI.Flash Thinking can also interact with Google apps like Maps, YouTube and Search."These connected apps already make the Gemini app a uniquely helpful AI-powered assistant, and we're exploring how new reasoning capabilities can combine with your apps to help you do even more," Patrick Kane, director of product management for the Gemini app at Google,wrote in a blog post.The news keeps comingThe wide release of Google Gemini 2.0 Pro Experimental, Flash-Lite and Flash Thinking is the latest in a long line of AI announcements from Google.Late last month, the Samsung Galaxy S25 series became the first smartphones to ship with Google's Project Astra, a feature where users can point their phone cameras at things and get AI input based on context. The company also launched its Daily Listen feature, which generates audio summaries of news updates based on your Search and Discover activity.For Google, it's a large string of rapid announcements to an industry currently riding high on DeepSeek, which boasts similar performance to bigger AI models at a fraction of the price. It's too soon to say if DeepSeek will stick around for the long haul due to a variety of factors, but Google's announcement and OpenAI's o3-mini release may help pry attention away from the young AI upstart.