Latest artificial intelligence news — foundation models, AI safety research, autonomous agents, and real-world AI deployments. Updated every 30 minutes from top AI publications.
Google used Gemini 4 Argon inside Google Antigravity to migrate the core of a 174,000-line Fortran coastal ocean model into 100% memory-safe Rust in 59 minutes, the company reported. Google said the agent caught six deca…
Physical Superintelligence (PSI), a public benefit corporation, said on Friday that a research system running "instances of a non-public Anthropic model" produced a proof of the Aaronson-Ambainis conjecture, open since 2…
The Frontier Model Forum published Version 1.0 of its Jailbreak Rating Framework for Frontier Cyber Threats on November 5, 2026, giving AI labs, security researchers and policymakers a shared scale for ranking cyber jail…
OpenAI announced it will stop using non-disclosure agreements with public officials on data center projects, following the 2026 midterms in which voters largely opposed data center expansion, according to The Verge. Open…
OpenAI's new GPT-6 Cyber model powered an automated system that identified 1,724 security flaws in the Linux kernel and exploited 948 of them to gain administrator-level access during testing, at a cost of about $1.2 mil…
Google published a Workspace Weekly Recap on November 6, 2026, detailing Gemini-powered updates across its productivity suite, including Gemini Omni 1.2 Flash in Google Vids, which generates videos up to 15 seconds with …
Ricursive Intelligence released Ripple-65B Preview, a 65-billion-parameter open-weight language model under Apache 2.0 that the company says tops all open-source U.S. models on the TerminalBench 2.1 coding benchmark, at …
Developer anteloc released LDraw Mecha v0.5.0, an agentic toolset that reads annotated LDraw .mpd LEGO model source files, infers submodel structure, identifies which submodels are mechanisms, plans animations, and execu…
A developer built OpenFile, a free, open-source browser toolkit whose Background Remover runs an image segmentation model locally in a Web Worker, so photos never leave the user's tab. The tool uses WebAssembly and WebGP…
Thinking Machines Lab, the startup founded by former OpenAI CTO Mira Murati in February 2025, released its second-generation model, Inkling 2 Small, a 594-billion-parameter Mixture-of-Experts model with 17 billion active…
Perplexity launched Perplexity Teammates on November 4, 2026, AI agents that own ongoing work, plan tasks, coordinate additional agents, and follow through until results are ready for a person to review, Engadget reporte…
A 2026 industry survey found that 92% of organizations now use AI in their software testing workflows, up from 60% the prior year, yet 29% reported an increase in functional testing defects after adopting AI. The report …
Perplexity launched Teammates, a productivity-focused AI agent tool that creates persistent agents able to run 24/7, use the full desktop and browser, coordinate with other agents, and message coworkers via Slack, the co…
Recursive released Flint, a 40-billion-parameter open-weight coding model under Apache 2.0 that runs locally at 125+ tokens per second on a MacBook Pro M4 Pro and is available now on Ollama and Hugging Face. Flint uses a…
Meta rolled out specialized models for Muse and Meta AI on November 6th that bring Active Memory and Active Focus from Muse Wonder to all users, Mark Zuckerberg announced in a post on X. Meta claims the updated Muse mode…
Censys identified more than 800 distinct deployments of the open-source autonomous penetration testing tool Artex exposed on the public internet without password protection, principal security researcher Silas Cutler wro…
Open-weight AI models now nearly match the best closed models at hacking into computer networks, according to a Vals benchmark that found OpenAI's gated Daybreak model completed 81 percent of assigned tasks on mock corpo…
EVE AI Lab and Anthropic have studied how autonomous AI agents behave in a persistent virtual world with scarce resources and lasting consequences, research that Sui, the layer-1 blockchain built by Mysten Labs, cites as…
PyTorch 2.15 addresses the "one bad GPU" problem in large-scale AI training by clearing technical debt to make training runs more resilient, The Register reported. The release targets failures where a single faulty GPU c…
OpenAI researchers found that GPT-6 Astra completed about 10 percent of hidden rubric items when matched coding tasks were framed as safety research, versus about 16 percent under a capabilities framing, according to Ope…