InterpretabilityResearchIn-context Learning and Induction HeadsMar 8, 2022Read PaperResearchPreparing for AI’s economic impact: exploring policy responsesOct 14, 2025ResearchA small number of samples can poison LLMs of any sizeOct 09, 2025ResearchPetri: An open-source auditing tool to accelerate AI safety researchOct 06, 2025