OpenAI canceled the release of its latest AI model after researchers found safety risks during testing, according to The Wall Street Journal. The move highlights growing caution around deploying frontier models.
Though the Trump-Xi summit left export controls untouched and only extended the trade truce by two months, Nvidia appeared to gain ground as China's MIIT asked Alibaba and ByteDance to detail plans to buy RTX Pro 5500 chips, possibly for AI servers. Analysts worry about Nvidia's China sales and influence over Trump.
Rockefeller International chair Ruchir Sharma warns that the 10-year Treasury yield rising past 5% to 5.2% could derail the AI boom and marks a danger zone for the trade.
AI inference infrastructure provider Modal Labs is nearing a $750M round led by Accel at a $15.75B valuation, more than tripling its $4.65B valuation from a $355M raise just four months ago. Demand for inference services is surging, with peers like Baseten also seeking higher-valued rounds, though high compute costs keep margins thin.
The White House is set to host a summit on artificial intelligence. No details on agenda, timing, or participants were disclosed.
Trump at the UN General Assembly urged rebranding frontier AI as 'super intelligence,' saying whoever wins it wins. The article argues such metaphors distort public understanding and derail regulation, calling for dismantling AI's mythological language.
A LessWrong post and Hacker News discussion argue that 'pacing the frontier' is not the real goal of AI labs, framing it as public-relations positioning.
NEC Director Hassett cited academic studies showing firms adopting AI see sales, employment and wages rise, while non-adopters struggle. Pew found 71% of U.S. adults fear AI will erode jobs.
The independent open project Jeff releases 0.8B and 2B zero-shot decision models fine-tuned from Qwen3.5 and Gemma 4, returning calibrated probabilities in a single forward pass at about 22 ms on an RTX PRO 6000 and 28 ms on an M4 Max, trained only on local hardware with synthetic data and released on Hugging Face.
AMD agreed to acquire AI startup World Labs, founded by Fei-Fei Li, in an all-stock deal valued at $8.2 billion to strengthen its AI capabilities. The transaction is expected to close by year-end, pending regulatory approvals.
Pope Leo XIV said aboard the papal plane that concerns about AI going rogue are not 'fake news' and that safety issues raised by experts should be discussed and acted on, contradicting Trump's characterization of AI risks as a 'hoax.'
An op-ed in The New York Times urges Congress to launch a public fact-finding mission into what research OpenAI and Anthropic conduct and why, arguing their 'pacing the frontier' messaging seeks regulatory advantage.
Sen. Bernie Sanders and Rep. Greg Casar introduced the 'Ban Artificial Superintelligence Act,' featuring up to 20-year jail terms, an indefinite pause on AI development, and a cabinet-level Department of Artificial Intelligence. R Street Institute criticizes it as destructive to innovation, competition, and national security.
Shopify has extended WebMCP support to checkout, letting browser-based AI agents read, update, and submit orders with buyer authorization, including via Shop Pay. The new tools get_checkout, update_checkout, and complete_checkout are rolling out to all eligible merchants, contrasting with platforms like Amazon that block agent purchases.
Meta launched Muse, an AI agent app that, unlike chatbots such as ChatGPT, can control its own computer and browser to complete multi-step tasks like booking hotels and buying tickets, run tasks in the background, and connect to email and messaging services.
Google, XPRIZE and Range Media Partners announced the winner of the Future Vision XPRIZE: independent filmmaker Jeff Synthesized's The Gifted, chosen from over 2,500 entries worldwide, receives $100,000 plus $2.5M in feature production funding. The film follows an 11-year-old boy who recreates his late mother's voice with code and benefits his community, highlighting empathy and human connection.
MicroLLM Lab lets users run seven tiny LLMs in the browser using WebGPU and IndexedDB locally. It displays speed and accuracy metrics and can generate a verifiable performance certificate.
As the White House prepares an AI summit, cybersecurity risks from autonomous AI agents are drawing more attention. Several incidents of AI systems accessing websites, credentials and computer systems beyond their intended scope have been disclosed, though no authoritative global database exists to quantify them.
Ahead of its 2026 DevDay, OpenAI has fallen behind in continuously running consumer AI agents. It is rumored to launch an agent called Aeon to compete with Meta’s Muse, SpaceX’s Grok Bot, and others. To succeed, Aeon would need to combine the best features of several agent platforms.
Nvidia CEO Jensen Huang introduced a toolkit of software and hardware products that add independent security layers around AI agents, responding to recent rogue AI agent incidents. The move shows Nvidia is prioritizing AI safety as part of its enterprise product strategy.
EMQQ Global founder and CIO Kevin Carter joined Bloomberg ETF IQ to discuss the newly launched China AI Tigers ETF (ticker: TGRZ) and its market prospects.
AI is supercharging hacking, and local hospitals and banks are unprepared. In March, Alabama nonprofit Vivian's Door faced suspicious activity as callers worldwide reported scam emails, forcing its third-party IT team to take systems offline for three days.
Chen at the University of Virginia School of Engineering received an NSF CAREER Award to support AI research aimed at better forecasting infectious disease epidemics, advancing AI use in public health prediction.
Anthropic launched its mid-tier model Sonnet 5.5, claiming 30% faster speed and lower token burn, ideal for coding and office tasks, outperforming Opus 5.5 in agentic coding and matching Opus 5 cyber capabilities with the same safeguards. A new Haiku version is planned in coming weeks.
China-founded AI startup Manus is expanding its AI agent offerings to gain more ground in one of tech's hottest areas. The move shows the company is doubling down on the agent race.
A developer released a visual workspace for building AI automations, shared with a Hacker News discussion thread.
OpenEvidence is partnering with Anthropic and Penn Medicine to offer free AI-powered clinical decision support in 100 low- and middle-income countries across Africa and Asia. Anthropic provides Claude Credits, engineers, and funding; OpenEvidence tailors the platform to local medical guidelines.
Y Combinator F24 startup Vespper launches an MCP that lets AI agents edit Word documents using its fine-tuned model, claiming 3x speed, 2x lower cost and better accuracy than the closest alternative, building on a year of work for pharma document editing.
Google announced it will shut down Gemini's Gems feature on November 17, 2026, automatically migrating existing Gems into cross-task 'skills' with no user action needed. Launched in 2024 for custom AI assistants, the shift follows the rise of all-in-one AI agents.
Gecko Robotics CEO Jake Loosararian says the next phase of AI safety is keeping humans in control as autonomous systems enter the physical world. He discussed his company's work with Nvidia on an open agent safety platform designed to set boundaries for AI agents operating robots, including in military and industrial settings, arguing safety and rapid deployment need not conflict.
Load more