AI research and discussion on LessWrong. www.lesswrong.com
Opinion piece arguing AI safety groups should prepare for a possible public anti-aging preference cascade that could fuel AI accelerationism, proposing policy roadmaps that narrowly scale AI for medicine while slowing general AI development.
Proposes pacing AI development, meaning interventions that moderate its pace, as a dedicated research field, covering incentives, regulatory and technical tools, and the full lifecycle of interventions.
Proposes a taxonomy for verifying and enforcing international AI pauses via compute monitoring, estimating that pauses are enforceable for training runs above roughly 10k H100-equivalents absent concealment.
Proposes how to slow AI progress if needed, arguing for layered compute caps and taxes on R&D and deployment, timed to critical capability windows while managing overhang risk.
Argues that a global AI pause is nearly impossible because treaties come after warning shots, AI is dual-use, there is no capability ceiling, defection is hard to detect, and no clear warning shot may occur.
Geoffrey Miller describes his NatCon talk arguing that superintelligence threatens extinction and social institutions, and urging conservatives to support a global ban on ASI development.
Analyses political pathways to slowing superintelligence development, including public worry over jobs, existential-risk-driven slowing, national security framings, and the role of warning shots.
A blog post proposing that AI safety advocates rapidly mobilise an outside-game movement and fund groups opposing AI companies in case of an AI investment crash.
Argues the AI safety community should prepare for a financial crash of AI labs, using the weakened industry to push for restrictions on AI companies and building a broader counter-movement.
Proposes and reviews a conditional international AI safety treaty with compute thresholds, mandatory model audits, and AISI-led pause authority to address loss-of-control risks from advanced AI.