Bilal Chughtai leaves DeepMind: 'AI has the potential to kill us all'
Chughtai actually left DeepMind back in July. He published his statement two months later — on the very day Trump called the industry's warnings a hoax.
Gemini 3.8 Live: Google lets the model talk while it is still thinking
97 languages it switches between mid-sentence, without being told. And tool calls that keep running in the background while the conversation carries on.
Compaction on demand: the Messages API summarises when you say so
The call returns no reply at all, just one signed block. It is still billed and rate-limited like any other request.
Musk wants the AI labs to test each other's models - China included
Not a single rival has signed up so far - Musk says so himself. The White House answered on Tuesday that the private sector is the right place for this.
Virologists Push Back on Anthropic's Bioweapons Report
Anthropic described five cases where researchers used Claude for sensitive work on viruses and toxins. A Scripps virologist says most of it is ordinary basic research.
Anthropic Is RUM Group's $13.7 Billion GPU Customer
The Information names Anthropic as the previously unnamed buyer of $13.7 billion in GPU capacity — one day after Dario Amodei floated limits on training compute.
Anthropic Expects a Second Straight Profitable Quarter
The Financial Times reported on Sunday that Anthropic told shareholders it will post an adjusted operating profit this quarter. Two words in that sentence deserve a second look.
Anthropic, OpenAI and Google Are Talking About Their Own Standards Body
The Information reports that the three largest labs have discussed a joint organization for AI safety standards. Which means the standards their own models would be measured against.