A LessWrong post proposes a two-pronged strategy for slowing down AI progress if catastrophic risks or societal instability emerge. The first prong suggests limiting AI R&D compute through a hard cap and a progressive tax, supplemented by a cap on individual training runs to mitigate sudden risks like misalignment. The second prong recommends a capability-gated tax on AI deployment to control the pace of economic displacement due to mass unemployment. The author emphasizes that these mechanisms should be dynamic, conditional, and targeted at specific layers of the AI tech stack corresponding to the nature of the risk. AI
IMPACT Proposes specific policy mechanisms that could influence the pace of AI development and deployment if implemented.
RANK_REASON The item is an opinion piece discussing potential policy mechanisms for AI slowdown, rather than a direct announcement or event.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
