EA Forum Bot Site
Topics
EA Forum

AI evaluations and standards

•

Applied to I read every major AI lab’s safety plan so you don’t have to 2d ago

•

Applied to OpenAI's o1 tried to avoid being shut down, and lied about it, in evals 12d ago

•

Applied to OpenAI's CBRN tests seem unclear 1mo ago

•

Applied to College technical AI safety hackathon retrospective - Georgia Tech 1mo ago

•

Applied to Comparing AI Labs and Pharmaceutical Companies 1mo ago

•

Applied to The current state of RSPs 1mo ago

•

Applied to Trendlines in AIxBio evals 2mo ago

•

Applied to Announcing ForecastBench, a new benchmark for AI and human forecasting abilities 3mo ago

•

Applied to Join the $10K AutoHack 2024 Tournament 3mo ago

•

Applied to Model evals for dangerous capabilities 3mo ago

•

Applied to Submit Your Toughest Questions for Humanity's Last Exam 3mo ago

•

Applied to Thinking About Propensity Evaluations 4mo ago

•

Applied to A Taxonomy Of AI System Evaluations 4mo ago

•

Applied to Case studies on social-welfare-based standards in various industries 6mo ago

•

Applied to [Paper] AI Sandbagging: Language Models can Strategically Underperform on Evaluations 6mo ago

•

Applied to Demonstrate and evaluate risks from AI to society at the AI x Democracy research hackathon 8mo ago

•

Applied to LLM Evaluators Recognize and Favor Their Own Generations 8mo ago

•

Applied to OMMC Announces RIP 9mo ago

•

Applied to Join the AI Evaluation Tasks Bounty Hackathon 9mo ago

•

Applied to Introducing METR's Autonomy Evaluation Resources 9mo ago