Anthropic's Safety Brand Cracks From the Inside Out
Anthropic researcher Jacob Coxon resigned over safety concerns as a senior colleague assigned 10% odds to AI killing all humans by decade's end.
Anthropic researcher Jacob Coxon resigned over safety concerns as a senior colleague assigned 10% odds to AI killing all humans by decade's end.
Cognition's $48B valuation arrived five days after OpenAI cut Cursor loose. The market priced infrastructure risk fast. Here's what the arc reveals.
Hackers are stealing Claude tokens from subscriber accounts. Anthropic warned users. No scope or mechanism disclosed. What the incident actually signals.
OpenAI launched ChatGPT Images 2.5 with a Sketch feature on Sept. 8, 2026. One test, one cat, one marketing headline. Here's what actually shipped.
Meta's Muse agent reaches email, calendars, payments, and health data — four classes the ad stack never had structured access to before.
Meta's Muse agent opens browsers, fills forms, and negotiates autonomously — moving from passive behavioral observation to active participation.
A class action filed Sept 8 alleges Anthropic misled Max subscribers. Two ex-FTC attorneys make the case. The gap between claim and delivery is on record.
OpenAI announced a "landmark" math discovery. No math is visible, no accusers are named. There's nothing to evaluate yet.
Google Cloud's Accenture deal targets enterprise AI's real bottleneck: operationalization, not compute. The thesis is testable. Watch what deploys.
Mistral raised €3B at €21B valuation. The capital is real; the "sovereign AI" framing around it is political and worth reading carefully.
Google DeepMind's AlphaGenome Atlas maps every possible DNA letter change across ~3B pairs. Real tool, large claim, blog-post sourcing.
Adobe's Generative Media tool puts existing AI generators inside Premiere's timeline. The capability isn't new — the friction cost is what changed.