Daily Briefing

OpenAI releases GPT-6.1 Terra with computer use capabilities and halved API pricingvia OpenAI
Fast, capable model for agent work at $1/$5 per million tokens, matching pricier models on key benchmarks.

OpenAI launches 'dots', an always-on personal AI assistantvia TechCrunch
OpenAI's coding tools evolved into a persistent agent product, though exec Alexander Embiricos says he never wants to meet with someone else's AI.

OpenAI announces three major breakthroughs in math, physics, and computer sciencevia OpenAI
Including that the Exponential Time Hypothesis, a longstanding conjecture, is false.

Apple's 20B on-device model runs in just 2GB of memoryvia AI Realist
Keeps most weights in flash storage and activates only 1-4 billion parameters per prompt.

AI secretly embedded across Home Office, creating 'digital hostile environment'via The Guardian
Open Rights Group report claims AI is now involved in tasks from asylum age determination to border surveillance.

Meta plans record $95 billion equity raise to fund AI expansionvia Reuters
Massive capital raise signals continued AI infrastructure spending from tech giants.

SpaceX closes $40 billion debt deal for AI chipsvia CNBC
Musk's space company secures GPU financing to expand AI capabilities.

Anthropic IPO seen as critical test for AI investment frenzyvia CNBC
Wall Street to Silicon Valley expecting massive 'trickle-down effect' from the offering's performance.

Inside Anthropic's massive—and wild—IPO prospectusvia Axios
There has never been a company like this: $650B in computing obligations, eye-popping revenue per employee.

The Anthropic S-1 filing is also an AI doomsday primervia WIRED
The prospectus doubles as a comprehensive guide to existential AI risks.

Economic policy vets see risks in AI overreliancevia Axios
Former White House economist Jason Furman warns about government dependency on AI systems.