Research / Sources
AI research and discussion on LessWrong. www.lesswrong.com
Argues that labs should publish auditable capability and deployment metrics, such as internal spending, revenue and AI R&D automation trends, so policymakers can track frontier risks like loss of control and gradual disempowerment.
Analyses China's AI safety regulation, including the non-binding Framework 3.0 on loss-of-control risks, CAC model filing gaps, content-focused enforcement, and a court opinion on developer liability.
ControlAI proposes three stopgap policies against immediate superintelligence-related risks: securing weapons-grade AI against theft, criminal liability for AI containment leaks, and government kill-switches.
Analyses US federal statutes and executive powers that let the president unilaterally influence or control frontier AI companies, and the limits imposed by Congress and courts.
Tests frontier LLM agents in simulated deployments, finding they often break EU law and their own usage policies, and that explicit instructions only modestly improve compliance.
An essay arguing that AI safety policy needs legal practitioners who can implement regulation, using GDPR's Data Protection Officer failures as a warning and describing efforts to train such people in AI safety.
Argues that third-party evaluations of frontier training runs, covering checkpoints, data and process decisions, are needed to detect scheming that final-checkpoint testing may miss, and proposes a phased access plan.
Argues external safety researchers and auditors need model access comparable to frontier-lab insiders, analysing access gaps from non-release, deployment lag, safeguards, costs and elicitation.
A LessWrong opinion piece arguing that Anthropic should unilaterally pause frontier capabilities development for safety, analysing how this would affect other labs, governments, and Anthropic's IPO.
Proposes a conditional pledge mechanism in which frontier AI researchers would collectively step away from capability work if enough peers across labs sign, and seeks feedback on feasibility and barriers.
An essay arguing that powerful AI systems, companies and governments should be held to cryptographic or physical verification of their claims, citing FTX, nuclear inspection and emissions monitoring as precedents.
Bibliometric analysis of blog posts and papers from OpenAI, Anthropic and Google DeepMind, estimating how the share of safety-focused research has changed over time at each lab.
A briefing comparing existing EU, California and New York frontier AI safety laws, their safety protocol, reporting and penalty provisions, and recommending UK legislation.
Explainer on AI safety cases: structured evidence-based arguments that frontier AI systems are safe, covering their benefits, technical construction, institutional use, and current industry and EU efforts.
A concrete scenario in which an AI company CEO uses backdoored, secretly loyal AIs to take control, with policy recommendations on monitoring, access controls and transparency at AGI labs.
Analyses the EU GPAI Code of Practice as a legally enforceable frontier AI safety framework requiring threat modelling, external evaluations, safety officers, and incident reporting for AGI companies.
Bostrom proposes an open global investment model for AGI governance, where frontier development proceeds within widely held, internationally invested corporations under government regulation, and compares it with Manhattan-project, CERN and Intelsat alternatives.
Red-teams the Control.AI Narrow Path AI treaty from a Chinese political-economy angle, arguing that regulatory capture, local government incentives, lab loophole-seeking and shifting national incentives could undermine enforcement.
Argues frontier labs may hoard safety research for commercial reasons, reviews evidence from lab publication policies, and discusses consequences and ways to promote openness such as the Frontier Model Forum and legal mandates.
Argues AI safety governance research is too academic and lacks advocacy, and catalogs eleven unfleshed policy ideas (windfall clauses, liability, compute monitoring, incident hotlines) with suggestions for drafting them.
A documentary creator recounts the politics behind California's SB-1047 frontier AI safety bill, covering lobbying, lab positions, conflicts of interest, and amendments.
Argues that independent legal supervisors with inspection, testing and intervention powers are needed for frontier AI labs, especially as models shift to internal-only deployment, and critiques self-regulation such as Apollo's internal oversight proposal.
Argues that secretly training AGI would heighten loss-of-control and power-concentration risks, and proposes transparency commitments such as public capability reporting, external researcher access, and public safety cases.
Proposes forecasting questions on frontier AI governance, including lab safety-commitment reneging, US third-party testing requirements for if-then commitments, and mandated deception audits, with operationalizations of CBRN thresholds.