OpenAI launches GPT-5.6, but only to companies the government clears first
OpenAI's most capable models yet shipped today as a tiny, government-vetted preview, signaling that Washington now holds a gate in front of the frontier.
The US government quietly lets Anthropic turn its most powerful model back on
Two weeks after ordering it switched off, Washington cleared Anthropic's Mythos 5 for release to more than a hundred trusted US institutions, a notable de-escalation.
Google DeepMind loses four senior scientists in six days, including a Nobel laureate
A Transformer co-author left for OpenAI and an AlphaFold Nobel laureate left for Anthropic, part of a fast run of senior departures that rattled Alphabet's stock.
DeepMind's plan for when an AI agent goes rogue: treat it like an insider threat
Google DeepMind published a defense-in-depth roadmap that assumes an AI agent might misbehave and uses a trusted supervisor AI to watch it in real time.
OpenAI launches Daybreak, an AI that finds and patches security holes for you
OpenAI's new cyber-defense program turns its models into an automated security team that prioritizes real threats, writes patches, and tests them, going head to head with Anthropic.
A huge study finds AI is more persuasive than trained, paid human experts
Across nearly 19,000 conversations, AI outargued incentivized human experts and raised real donations far more effectively, but its edge collapsed when slowed to human speed.
Frontier AI is getting more expensive while open models keep getting cheaper
Closed frontier models are raising prices and tightening access just as Chinese open-weight models slash theirs, a structural reversal with big consequences for who builds with AI.
Why making an AI think out loud helps it remember facts, even nonsense thinking
Google Research found that reasoning traces help a model recall facts partly just by buying it extra computation, so even repeating 'let me think' helps, though hallucinated steps backfire.
Why teaching AI agents to use tools keeps blowing up in training
A new paper pins the sudden collapse of multi-step tool-use training on runaway probabilities in a few control tokens, and shows that mixing in supervised examples stabilizes it.
Big Tech is set to spend up to three-quarters of a trillion dollars on AI in 2026
Projected AI infrastructure spending for 2026 runs into the hundreds of billions, financed increasingly with debt, as OpenAI also moves into custom chips to cut inference costs.