Slicast
Thursday, October 22, 2026
DarkSubscribe
Slicast
AI Infrastructure · News & Analysis
Home › Policy › Report
Policy · Report

AI safety advocates are calling on OpenAI and Anthropic to cap the compute they devote to internal R&D, according to The Information.

If adopted, such a cap would constrain internal research compute demand at leading labs and could become a governance benchmark for frontier-model capex.
Trade pressSlicast · October 21, 2026 at 14:00 UTC · US · Source: The Information
importance 40
Read the original
Companies in this story
OpenAI
Anthropic
In-depth analysis
CommentaryOpenAI on Google Cloud, October 2026: Microsoft Exclusivity Ends as $20 Billion Revenue Gap Tests the BuildoutCommentaryOpenAI Annualized Revenue, October 2026: $50 Billion Actual vs. $70 Billion Previously Cited
Related reports
PolicySam Altman reportedly laid out three endgames for AI safety in an internal OpenAI presentation: an international watchdog modeled on the IAEA, a merger of the major labs, and a scenario in which one firm 'melts everyone else's GPUs'.2026-10-21PolicyAnthropic's proposed AI safety bill would apply only if 50 deaths or $1 billion in damages are at stake, and would require evaluators to report imminent risks to the Department of Commerce within 72 hours.2026-10-19PolicyOpenAI, Nvidia, Anthropic and others have pledged $2.4 billion in computing resources to Trump's Genesis Mission.2026-10-11PolicyOpenAI and Anthropic are hiring a former Trump administration official, according to Analytics Insight.2026-10-11PolicyChina's DeepSeek and peers release 16 AI foundation models in one month despite U.S. export controls on cutting-edge GPUs, signaling adaptive supply-chain resilience.2026-10-07
AI safety advocates are calling on OpenAI and… · Slicast