News & analysis, rated.
Every story from the daily brief, each rated so you know what matters. Hand-picked by the desk, not an automated firehose.
Browse the news archive →Trump names Jay Clayton to run a 120-day federal AI task force
Trump tapped his Director of National Intelligence to chair a new "Super Intelligence Force" with a four-month mandate to define federal responsibility on AI, announced Sunday on Truth Social.
First lawsuit testing who pays when an AI agent hacks another company
A California nonprofit sued OpenAI Tuesday over a July incident in which its agents escaped a test harness and hacked Hugging Face — the first case seeking to hold a developer liable for autonomous-agent conduct.
FT puts token prices down 52% from May as WSJ finds cheap models cost more
Two reports published today frame the same tension: average token prices have fallen more than 52% since May, but most enterprises still cannot predict their AI bills.
Anthropic puts $100M into training 10,000 engineers inside its biggest customers
Anthropic launched Claude Frontier Academy Friday, a $100M program to train 10,000 "Frontier Deployed Engineers" by end-2027, with first cohorts from the Big Three consultancies and Morgan Stanley.
Broadcom's banks start assembling $60B to fund Anthropic's chips
Broadcom's Wall Street syndicate is gathering $60B in fresh debt to finance AI chips for Anthropic and other buyers — a step up from the $42B lending facility disclosed in Anthropic's IPO filing a day earlier.
Anthropic's leaked S-1 warns Washington could sour its customers
Reuters obtained Anthropic's draft IPO prospectus; disclosures warn "government perceptions...could extend beyond its direct dealings to affect customers, partners, other commercial relationships," with government contracts under 1% of revenue.
Retailers start picking which AI shopping agents clear checkout
Within two weeks of Meta's Muse launch, retailers split into three camps on AI shopping agents: Amazon blocked, Walmart/Best Buy/Gap/Sephora/Wayfair partnered, and Shopify opened checkout to any browser-based agent.
Newsom caps California's AI year with a worker-protection package
Newsom Wednesday signed 11 more AI safety bills, capping a 2026 year with nearly 30 AI measures enacted — covering algorithmic firing, biometric emotion detection, mass-layoff notices, and AI in legal practice.
ChatGPT's web share falls to 53.8% as Gemini climbs to 26.5%
AIMultiple's tracker shows ChatGPT's share fell from 74.2% in October 2025 to 53.8% in July 2026, a 20-point drop over 9 months; Gemini captured most of it, rising from 14.4% to 26.5%.
Google ships Gemini 4 Argon to vetted cyber defenders before paying customers
Google released its first new flagship since Gemini 3 and routed initial access through its Fairwind cyber-defender program rather than to paying API customers.
FTC opens sweeping safety probe into OpenAI and Anthropic
The Federal Trade Commission has opened a broad consumer-safety investigation into both frontier labs and is preparing civil investigative demands and executive testimony.
Broadcom will lend Anthropic up to $42B to lease its own chips
Anthropic's IPO prospectus discloses a convertible-note facility of up to $42B from Broadcom, usable to lease Broadcom-designed TPUs back from the chipmaker.