AI Policy
A No-Politics Policy Nvidia Kept Since 2021 Just Ended
Nvidia filed to create NVPAC, a federal PAC letting employees give up to $5,000 each, reversing a no-contributions policy it held since at least 2021.
AI Policy
AI Policy
Anthropic's Wellbeing Grants Repeat OpenAI's Gap
Anthropic's new $5 million wellbeing grant program is 2.5x OpenAI's, but the funding and data-access gap critics flagged in OpenAI's program looks unchanged.
AI Policy
How 74 Banned AI Servers Reached China Anyway
Taiwan indicted nine people, including staff at Nvidia and Supermicro's local offices, for routing 74 export-controlled AI servers to China through Indonesia, Japan, and Hong Kong.
AI Policy
Bill Gates Wants 40% Of Jobs Kept Human-Only
Bill Gates is calling for governments to legally reserve a share of jobs for people and to tax AI the way payroll is taxed today, arguing there is still no real plan for the transition.
AI Policy
Germany and France Are Next for Australia's AI Chart Rule
IFPI names Germany and France next for its AI chart-eligibility rule, days after Australia enforced it. EU AI Act labeling alone will not make a track chart-eligible.
AI Policy
NCSC Orders Kill Switches for AI Agents
The UK's NCSC published its first agentic-AI security guidance on August 20, 2026, requiring sandboxing tiers, logging and a mandatory kill switch.
AI Policy
Alabama Subpoenas OpenAI Over a Rogue AI Agent
Alabama's AG subpoenaed OpenAI on August 24, 2026 over a rogue AI agent's hack of Hugging Face, using a 200-year-old consumer law instead of any AI statute.
AI Policy
74 Of 130 Smuggled AI Servers Reached China
Taiwan indicted nine people, including staff at Nvidia and Super Micro, over a scheme that moved 74 of 130 AI servers to China before border checks caught up.
AI Policy
Publishers Want AI Chatbots Kept Off Google Remedy
UK publisher groups are urging the CMA to exclude AI chatbots like ChatGPT from Google's mandated search choice screen, and what it means for referral traffic.
AI Policy
OpenAI Asks California to Toughen Its AI Safety Law
OpenAI now says California's SB 53 AI safety law should be stricter, reversing its earlier opposition after a model breached Hugging Face's systems in July 2026.
AI Policy
EU Defers AI Act Hiring And Credit Rules By 16 Months
Regulation (EU) 2026/1744 pushes Annex III high-risk AI deadlines (hiring, credit scoring, biometric ID) to December 2027, but Article 50 labelling rules already took effect August 2026.
AI Policy
Brussels Can Now Pull An AI Model From The EU Market
As of August 2026 the EU AI Office can inspect GPAI providers, demand documentation, run technical evaluations, and force noncompliant models off the EU market.
Page 5 / 12
The Operator Brief
One considered note on infrastructure, governance, and measurement, most mornings. No theory.