Daily intelligence on AI, technology, and the decisions that follow.
Written for people who make decisions — not people who make predictions.
Get The Signal →Free to start. No spam. Unsubscribe anytime.
New evidence shows multi-hour agent task horizons and widening control gaps. It still does not establish recursive self-improvement or economy-wide job replacement.
Read the feature →The Anthropic ban didn't end the safety-vs-military debate. It settled it — by making clear that in Washington, compliance wins contracts and principles cost them.
📡 Signal Brief — SB-2026-03-02 Monday, March 2, 2026 Nvidia Bets $4B on Photonics — Lumentum and Coherent Deals Signal Next Infrastructure Layer Infrastructure / Semiconductors Nvidia announced $4
📡 Signal Brief — SB-2026-03-01 Saturday, March 1, 2026 Anthropic Walks Away from the Pentagon — OpenAI Steps In Within Hours AI Governance / Defense Policy Negotiations between Anthropic
Block cuts 4,000 jobs citing AI, Anthropic refuses Pentagon contract, Claude used to breach 10 Mexican agencies.
Anthropic acquires Vercept, Amazon's $50B OpenAI deal terms revealed, AI war games end in nuclear strikes.
Nvidia $68B Q4, DeepSeek V4 locks out US chipmakers, Alibaba's Qwen3.5 matches Sonnet on 32GB hardware.
Not hype. Not fear. A timeline of the 13 months that rewired the AI stack — from the reasoning threshold to the first confirmed AI-assisted government breach.
Anthropic drops its safety pledge, Claude finds 500 zero-days, Liquid AI ships a 32GB frontier model.
GPT-4o is dead. Alibaba goes agentic. China's wave breaks.
Claude gets agent teams. OpenAI locks down. Hackers target agents.