News

London announced a new £500 million (about $675 million) Sovereign AI fund aimed at accelerating homegrown artificial‑intelligence companies. Led by venture‑capital partners James Wise and Joséphine Kant, the fund will back startups across model development, agentic AI and drug discovery, while granting them access to the nation’s supercomputing resources, visa shortcuts and procurement pipelines. The first tranche includes an investment in processor‑coordination firm Callosum and compute credits for seven other firms, signaling Britain’s push to become an AI maker rather than a taker. Read more

OpenAI unveiled a substantial update to its Codex AI coding platform, adding multi‑app agents, a built‑in browser, image generation, and early memory functions. The enhancements let developers command specific desktop programs, integrate 111 new plugins, and receive proactive suggestions. The rollout begins with macOS users logged into ChatGPT, with EU and UK releases slated for later. While the full super‑app that merges ChatGPT, Codex and a web browser remains in development, the latest release offers a tangible glimpse of OpenAI’s broader vision for a unified desktop AI experience. Read more

Perplexity announced today that its new Personal Computer AI assistant is now available for Mac users. The tool builds on the company's multi‑model orchestration technology first shown in Perplexity Computer and joins the ranks of Anthropic's Claude Cowork and OpenAI's Codex. Personal Computer can read and act on to‑do lists, organize files, interact with apps like Notes and Messages, and be controlled by voice or a smartphone. The rollout begins with Max subscribers, with broader access slated for wait‑list members. Read more

OpenAI has unveiled GPT‑Rosalind, a large language model tuned specifically for biology. The new system aims to curb the over‑enthusiasm and sycophancy that have plagued earlier models, offering more skeptical, fact‑checked responses on drug targets and other scientific queries. Access is restricted to U.S. entities through a trusted‑deployment program, with a broader Life Sciences Research Plugin slated for later release. OpenAI cites safety concerns, including the risk of the model being used to optimize harmful viruses, as the reason for the limited rollout. Read more

Google has added Nano Banana-powered image generation to Gemini’s Personal Intelligence feature, allowing the AI to create pictures that draw on a subscriber’s Gmail, Photos, Calendar, Drive and other Google apps. The capability launches this week for Plus, Pro and Ultra users in the United States, with free accounts slated to receive access in the coming weeks. Europe is excluded from the initial rollout due to regulatory concerns. Google says the tool does not train on personal data, but it does process that information to produce context‑aware visuals. Read more

Mike Krieger, chief product officer of AI lab Anthropic, resigned from the board of design platform Figma on April 14, the same day the company filed a notice with the SEC. The move follows a report that Anthropic’s upcoming Opus 4.7 model will embed design capabilities that could rival Figma’s core offering. Krieger, a co‑founder of Instagram and the AI news app Artifact, joined Figma’s board less than a year ago. Investors are watching closely as the potential clash fuels worries about a "SaaSpocalypse" in the software sector. Read more

AI infrastructure startup Upscale AI is reportedly in talks to raise between $180 million and $200 million, a move that would lift its valuation to roughly $2 billion. The company, founded just seven months ago, has already secured a $200 million Series A and a $100 million seed round, attracting investors such as Tiger Global Management, Xora Innovation and Premji Invest. Although Upscale AI has yet to launch a product, it is focusing on custom AI chips and the surrounding infrastructure, betting on a full‑stack, open‑standard approach to meet growing demand for scalable AI hardware. Read more

OpenAI announced a sweeping upgrade to its Codex coding assistant, adding background desktop control, memory recall, image generation and a suite of new plug‑ins. The changes let the tool run parallel agents on a user’s Mac, browse the web, and integrate with over a hundred apps. The move, positioned as a direct challenge to Anthropic’s Claude Code, also introduces a pay‑as‑you‑go pricing tier for enterprise customers. Read more

Factory, a San Francisco‑based startup that builds AI agents for enterprise engineering teams, announced a $150 million Series B round that values the company at $1.5 billion. The financing was led by Khosla Ventures with participation from Sequoia Capital, Insight Partners and Blackstone, and brought Keith Rabois onto the board. Founder Matan Grinberg said the firm’s edge lies in its ability to toggle between foundation models such as Anthropic’s Claude and DeepSeek. Customers already include Morgan Stanley, Ernst & Young and Palo Alto Networks. Read more

Sequoia Capital announced a $7 billion raise for a fresh fund aimed at late‑stage investments in artificial‑intelligence companies across the United States and Europe. The capital, nearly double the size of the firm’s 2022 vehicle, underpins Sequoia’s “expansion strategy” under its newly appointed co‑stewards Alfred Lin and Pat Grady. Backing already includes OpenAI, Anthropic and a slate of emerging AI startups, positioning the Silicon Valley stalwart to ride the rapid scaling of AI‑driven businesses. Read more

Anthropic announced it will occupy a new 158,000‑square‑foot office in London, enough to house up to 800 employees—four times its current head count. The move aims to deepen the company’s research and commercial presence in Europe amid a talent race with other AI labs. The expansion comes as Anthropic faces a legal dispute with the U.S. Pentagon over its refusal to allow its models in mass‑surveillance or weapon systems, while the U.K. government seeks closer cooperation on AI safety and security. Read more

Anthropic has begun rolling out identity verification for users of its Claude chatbot, requiring a government‑issued photo ID and a selfie in limited cases. The verification is handled by third‑party Persona, whose investors include Peter Thiel’s Founders Fund. While the company says the step targets fraudulent or abusive activity, many subscribers balk at the added biometric check, citing privacy concerns and the service’s ties to government surveillance firms. Anthropic maintains the data will be encrypted, not stored, and will never train its models. Read more

Anthropic has unveiled Claude Opus 4.7, the company’s most capable model offered to the public to date. Marketed as a step up from Opus 4.6, the new system promises stronger performance on software‑engineering tasks, improved image analysis, and more creative output for slides and documents. While Anthropic continues to restrict its flagship Mythos Preview to a handful of partners, Opus 4.7 ships with added cybersecurity safeguards and the same token‑based pricing as its predecessor. Early adopters include Intuit, Shopify, Databricks and other tech firms eager to test the model’s enhanced capabilities. Read more

New York‑based Antioch, a simulation platform for robot developers, announced an $8.5 million seed round that values the company at $60 million. Led by A* and Category Ventures, the round also includes MaC Venture Capital, Abstract, Box Group and Icehouse Ventures. Antioch’s technology aims to shrink the "sim‑to‑real" gap that hampers autonomous systems, letting engineers train robots in high‑fidelity virtual warehouses instead of costly physical testbeds. The funding will accelerate product development and expand the startup’s customer base, which already includes several large multinationals. Read more

Google has expanded Gemini’s Personal Intelligence feature to let the AI draw on users’ Google Photos libraries when creating images. Subscribers to Gemini AI Plus, Pro or Ultra in the United States can prompt the system with requests like “Design my dream house” and receive visuals that reflect their personal tastes and lifestyle. The capability, powered by the Nano Banana 2 model, identifies people and objects in a user’s photo collection to tailor the output, while Google says it will not train its core models directly on private images. The rollout begins in the next few days on Chrome desktop and will broaden to more users soon. Read more

Google announced Thursday that its Gemini Personal Intelligence feature will soon generate images using a new Nano Banana‑powered engine. The upgrade lets the AI create pictures that reflect a user’s preferences and photo‑library labels without explicit prompts. Subscribers to Google’s Plus, Pro and Ultra plans in the United States will receive the capability within days, and the company says it will roll out to Chrome desktop and other markets soon. The move expands Gemini’s contextual understanding, but Google warns the system can still misinterpret data and invites user feedback. Read more

A new analysis by Open Resource Applications shows Google Gemini’s output slips past popular AI‑detection tools more often than rival models, including ChatGPT and Grok. Researchers fed a dozen AI systems the same writing prompt and ran the results through Grammarly, QuillBot and GPTZero. Gemini registered the lowest detection rates, eluding Grammarly and QuillBot entirely while still tripping GPTZero’s stricter algorithms. The findings highlight growing uncertainty for educators, publishers and anyone relying on detection software to separate human‑written text from machine‑generated content. Read more

Researchers at the University of California, San Diego have unveiled GOFlow, an artificial‑intelligence system that extracts surface‑current information from thermal satellite imagery. Published in Nature Geoscience on April 13, the study shows the AI can track small, fast‑changing ocean currents with far greater resolution than traditional methods. By comparing GOFlow’s outputs to ship‑based measurements and conventional satellite data, the team demonstrated its accuracy while highlighting limitations such as cloud cover. The code will be released publicly, promising broader use of AI in Earth observation and climate research. Read more

Amazon Web Services unveiled Amazon Bio Discovery, an AI‑powered suite that lets researchers design, test and refine drug candidates in weeks instead of months. The platform links more than 40 specialized models with a lab‑in‑the‑loop workflow, letting scientists send top molecules to partner labs for synthesis and receive real‑time feedback. Early adopters such as Memorial Sloan Kettering Cancer Center, Bayer and the Broad Institute report dramatic cuts in development timelines, while AWS promises to keep scientists in the loop rather than replace them. Read more

Elon Musk’s SpaceX and Tesla groups have contacted semiconductor equipment giants Applied Materials, Tokyo Electron, Lam Research and Samsung Electronics for price quotes and delivery timelines on tools needed to build a U.S. AI chip fabrication complex dubbed Terafab. Bloomberg says the outreach reflects Musk’s “light‑speed” ambition to create a vertically integrated supply chain that could rival TSMC, Samsung and Intel, moving the project beyond concept into early procurement. Read more