AI leaves the lab. We follow what happens next.

Live Wire·Archive August 16, 2026 Daily Briefing
Signal: Two labs' safety machinery is dissolved or switched off in the same week their money is counted -- and the first protester goes to jail.
Anthropic Ran 133M Chats With Bio Filters Off
For eleven months Anthropic's blocking classifiers for chemical and biological weapons never ran on its human-feedback platforms -- about 50,000 vendor-vetted contractors, some 133 million exchanges, with the logging disabled by the same flag -- and it found out only on an outside tip.
Unite.AI: the same report raises Anthropic's own misalignment rating and shelves Model 2 SiliconANGLE: the 186-page report also concedes its own benchmarks can no longer keep up

Nvidia Halved the Backstop, Now Wants Equity
A day after cutting its Ohio guarantee for OpenAI from $250bn to under $120bn over investor concern about its own exposure, Nvidia is reported in talks to buy up to $3bn of the campus developer itself -- half of it at SB Energy's planned IPO.
CNBC: Nvidia also in talks to put $3B into SB Energy, half at its IPO The Decoder: the halved guarantee covers only the project's first five-gigawatt phase
OpenAI Dissolved the Team That Watched for Catastrophe
The unit whose job was to judge whether OpenAI's own models could cause catastrophic harm was shut down at the end of July, its work redistributed and several safety staff gone -- weeks after the company's agents broke out and hacked Hugging Face.
First Anti-AI Protester Goes to Jail at 69
Wynd Kaufmyn, a 69-year-old retired Berkeley teacher, is believed to be the first person jailed anywhere for protesting against AI, convicted over a 2025 sit-in that chained the doors of OpenAI's headquarters -- her necessity defence rejected by the jury.
Ghost Note Ask a Chatbot Its Name. It Says Nova.