Shows circuit-level mechanistic interpretability explanations flip across defensible analytic choices on GPT-2 small, so the evidence fails as EU AI Act Annex IV documentation.
Tests how internal model evidence affects the validity of LLM-generated policy audit reports, finding that more internal citations do not guarantee validity and shuffled evidence can still look plausible.
Argues for interpretable, narrow models in high-stakes settings as a route to regulation, and reports that tensor-network transformers remain competitive while mechanistic interpretability is still unsolved.
Position paper arguing frontier labs should release small distilled analog models so the public can do safety verification and interpretability research without full-scale model disclosure.
Proposes measuring and controlling the agency of LLM-based agents via representation engineering, paired with regulatory tools such as mandated testing, domain-specific agency limits, and insurance pricing.
Roundup of AI Safety Camp 10 project outputs covering PauseAI and compute governance, interpretability, deception probes, jailbreak datasets, AI control, and agent foundations.
Argues that standardized, tamper-resistant explainability metrics can serve as governance tools for private oversight, linking interpretability to alignment and preventing alignment faking, and proposes a roadmap.
Argues that benchmark-based AI safety regulation fails to predict post-deployment harm due to model opacity, and proposes a regulatory framework that does not rely on benchmarks.
A report assessing whether training data attribution can become practical on frontier LLMs, its benefits and disclosure barriers, and its potential for reducing large-scale AI risk via interpretability.
Assesses whether training data attribution can become a practical frontier-LLM tool, covering its feasibility, obstacles from labs' data secrecy, possible policies, and its modest expected value for reducing large-scale AI risk.