Bipartisan bill would force AI companies to build a kill switch
Reps. Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act on July 23, which would let Homeland Security order large AI companies to shut down a system after a serious incident and preserve its weights for audit.
Safety institutes measure Kimi K3's hacking ability: better than any open rival, nowhere near the top
The UK and US AI safety institutes found Kimi K3 scored 32% on an exploit-development benchmark versus 24% for the previous open leader, but reached working code execution in zero of 41 attempts where leading closed models average about half.
NeurIPS runs a monitored AI-review experiment and formally bans prompt injection in papers
NeurIPS released 2026 paper reviews on July 22 under an opt-in AI-assistance experiment, with a handbook that explicitly prohibits prompt injection and admits it cannot police prose merely tuned to please an AI reviewer.
Anthropic doubles its policy-advocacy funding to $40 million
Anthropic said on July 21 it gave a second $20 million to Public First Action, bringing its total to $40 million for a bipartisan nonprofit that it says is barred from spending on any candidate election.
Black Forest Labs launches FLUX 3 -- image, video, audio, and a robot that never renders the video
Black Forest Labs announced FLUX 3 on July 23 as a unified image, video and audio model with 20-second video and native audio, available only in early access, with open weights deferred to a later Dev release.
Ant's Ling-3.0-flash goes live free: 124 billion parameters, 5 billion doing the work
Ant released Ling-3.0-flash on July 23, a 124-billion-parameter model that activates only about 4% of itself per token, with a 256,000-token context and free access on OpenRouter and Vercel's AI Gateway.
AMD and Cerebras split AI inference across two different chips
AMD and Cerebras announced a joint inference offering on July 23 in which AMD's Helios racks process the prompt and Cerebras's wafer-scale engine generates the tokens, claiming up to five times the tokens per watt of a Cerebras-only setup.
Anthropic plans up to two gigawatts of AMD chips, with AMD committing up to $5 billion back
AMD said Anthropic plans to deploy up to 2 gigawatts of MI450-series capacity starting in the first half of 2027, and that AMD has committed to a future equity investment of up to $5 billion in Anthropic.
A newly minted Fields medalist says he is joining OpenAI's safety division
Jacob Tsimerman, awarded a 2026 Fields Medal on July 23, told journalists the same day that he will soon start a position in OpenAI's safety division, according to AFP.
Giving experts no memory cuts optimizer state from 50 gigabytes to 1.3
A single-author preprint shows that a mixture-of-experts model can drop momentum entirely for its expert layers, shrinking persistent optimizer state from 50.55 gigabytes to 1.29 with almost no effect on final quality.
A Huawei-chip training report shows what leaving CUDA actually costs
SLAI's technical report documents full-parameter post-training of a DeepSeek-V4 model on Huawei Ascend hardware, and the work list -- rebuilt collectives, converted checkpoints, hand-written kernels -- is the real measure of chip independence.
Three big AI layoff stories, and what the filings actually say
Oracle's headcount fell 21,000 in a year during an AI-linked restructuring, but its filings disclose no per-job AI attribution, Patreon's CEO explicitly denies AI replaced staff, and Meta reorganized around agents before admitting they were slower than expected.