AI Risk
AI risk research focuses on identifying, assessing, and mitigating the potential harms stemming from artificial intelligence systems, encompassing issues from bias and misuse to catastrophic failures. Current research emphasizes developing frameworks for risk assessment and management, often leveraging large language models and other advanced AI architectures to analyze risks and evaluate mitigation strategies. This work is crucial for establishing responsible AI development and deployment practices, informing policy decisions, and ensuring the safe integration of AI into society. The development of standardized risk assessment tools and the integration of safety and security considerations are key areas of ongoing focus.
Papers
From Transcripts to Insights: Uncovering Corporate Risks Using Generative AI
Alex Kim, Maximilian Muhn, Valeri Nikolaev
Managing extreme AI risks amid rapid progress
Yoshua Bengio, Geoffrey Hinton, Andrew Yao, Dawn Song, Pieter Abbeel, Trevor Darrell, Yuval Noah Harari, Ya-Qin Zhang, Lan Xue, Shai Shalev-Shwartz, Gillian Hadfield, Jeff Clune, Tegan Maharaj, Frank Hutter, Atılım Güneş Baydin, Sheila McIlraith, Qiqi Gao, Ashwin Acharya, David Krueger, Anca Dragan, Philip Torr, Stuart Russell, Daniel Kahneman, Jan Brauner, Sören Mindermann