Google Offers UK Teachers Free AI Training. Anthropic Trains Consultants
Google opened its AI Educator Series to all 650,000 educators in the UK and Ireland on Oct. 5. Anthropic's US$100m academy, announced…
AI & ML
Frontier models, the labs that build them, and the benchmarks, system cards and safety evaluations that decide what they can do. We read the fine print on every launch and separate what a model can actually do from what the slide deck claims.
42 stories
OpenAI Will Watermark ChatGPT Text in the EU. Why Britain Isn’t Getting It
OpenAI's invisible textGrain watermark spots about 95% of 400-token ChatGPT passages, falling to 17% when a quarter of the words are swapped.…
Francis Halzen Wins the 2026 Physics Nobel for a Telescope Made of Ice
The Wisconsin physicist takes the whole 12 million kronor prize for IceCube, a cubic kilometre of Antarctic ice that caught neutrinos from…
DeepSeek V4.1 Flash Cut the US AI Lead to 3%. What LiveBench Measures
Bloomberg Intelligence says DeepSeek's cheap, open-weight model trails Anthropic's best by 2.3 points on LiveBench, and beats it at agentic coding. Here…
Gemini 4 Argon, Explained: Why Google’s Top Model Went to Defenders First
Google's new flagship writes up to 1 million tokens in one answer and, by Google's account, finds and patches critical bugs on…
Sam Altman Says Accept ‘Some Bad Things’ From AI. Here’s the Full Context
The OpenAI CEO's line came from a Politico interview published Oct. 4, a day after his former safety lead quit in public.…
Will AI Win a Nobel Again? The Prediction Lists Say Probably Not
The physics and chemistry prizes land Oct. 6 and 7 at 5:45 a.m. ET, two years after neural networks swept both. Clarivate's…
What Is an AI ‘Safety Case’? How Labs Decide a Model Is Too Risky to Ship
Safety cases, red teams, outside evaluators and the new White House audit pledge, explained through two real tests: OpenAI's cancelled GPT-6.1 Astra…
OpenAI’s Safety Lead Quit Over Culture. California’s AG Was Already In
David Robinson, who led the drafting of OpenAI's Preparedness Framework, says the lab still learns safety by trial and error. A state…
The New AI Models Don’t Talk. They Decide.
Cloudflare, Perplexity and Amazon's Strands all shipped "decision models" on Oct. 1. They return probabilities instead of prose, and all of them…