Satinder Singh BavejaCo-principal investigator at launch
Academic center · Science policy · UC Berkeley, Berkeley · Est. 2016
The Center for Human-Compatible Artificial Intelligence (CHAI) is a multi-institution academic research center based at UC Berkeley that works on the theory, practice, and policy of “beneficial” AI—AI that is designed to be provably aligned with human values rather than merely capable of optimizing some arbitrary objective. CHAI’s mission is to develop the conceptual and technical foundations to reorient mainstream AI toward provably beneficial systems, explicitly connecting AI technical research with human and social science questions about what “beneficial” means.
It operates as a hub for a network of Berkeley and non-Berkeley scholars and collaborators across computer science and related fields (e.g., control, logic, economics/game theory, cognitive science/psychology, and policy).
No curated programs yet. Check the careers page ↗.
Khanh Nguyen, Benjamin Plaut, Tu Trinh, and Mohamad Danesh introduce a fundamental coordination problem called Learning to Yield and Request Control (YRC), where the objective is to learn a strategy that determines when to act autonomously and when to seek expert assistance. They
Computational Frameworks for Human CareBrian Christian, CHAI Affiliate, has published an article titled “<a href="https://www.amacad.org/sites/default/files/publication/downloads/daedalus_wi25_12_christian.pdf">Computational Frameworks for Human Care</a>” in the most recent issue of Daedalus, the journal of the Americ
A Practical Definition of Political Neutrality for AIThere is an urgent need for a clear, consistent, and practical definition of political neutrality for AI systems.
RvS: What is Essential for Offline RL via Supervised Learning?Scott Emmons, PhD student, was an author on “RvS: What is Essential for Offline RL via Supervised Learning?”
Getting By Goal Misgeneralization With a Little Help From a Mentor“Tu Trinh, Ben Plaut, Khanh Nguyen, and Mohamad Danesh wrote the paper, “Getting By Goal Misgeneralization With a Little Help From a Mentor.” This paper explores whether goal misgeneralization can be mitigated by allowing an agent to ask for help when it is uncertain. The answer
Linear Probe Penalties Reduce LLM SycophancyVisiting ETH MsC student Henry Papadatos and supervising CHAI PhD student Rachel Freedman publish an article “Linear Probe Penalties Reduce LLM Sycophancy” at the NeurIPS SoLaR workshop. The paper demonstrates a generalizable methodology for reducing unwanted LLM behaviors that a
Rachel Freedman selected as inaugural Cooperative AI FellowRachel Freedman, PhD Student, has been selected as one of the fellows for <a href="https://www.cooperativeai.com/phd-fellowship/2025">Cooperative AI’s PhD Fellow Program</a>.
Representative Social Choice: From Learning Theory to AI AlignmentTianyi Qiu, CHAI Intern, wrote this paper which was accepted by NeurIPS 2024 Pluralistic Alignment Workshop. <a href="https://arxiv.org/pdf/2410.23953">Here</a> is the link to the paper.
Getting By Goal Misgeneralization With a Little Help From a MentorKhanh Nguyen, Mohamad Danesh, Ben Plaut, and Alina Trinh wrote <a href="https://arxiv.org/pdf/2410.21052">this paper</a> which was presented at Towards Safe & Trustworthy Agents Workshop at NeurIPS 2024.
Language-Guided World Models: A Model-Based Approach to AI ControlKhanh Nguyen, CHAI Postdoctoral Fellow, published a paper at the Fourth International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics (ACL 2024).
“The Alignment Problem” Wins Xingdu Book AwardBrian Christian’s book “The Alignment Problem” was announced as the sole winner in the New Knowledge Category for Imported Editions at the Xingdu Book Award ceremony in China. The Chinese translation was published this past year by Hunan Science & Technology Press.
Social Choice Should Guide AI Alignment in Dealing with Diverse Human FeedbackRachel Feedman, CHAI Phd Student, and Wes Holliday, CHAI Affiliate, published a paper at the International Conference on Machine Learning
CHAI’s workshop site announces the dates and describes the 10th annual workshop as a Chatham House Rule retreat to examine emerging AI safety research questions.
Learning to Coordinate with ExpertsCHAI researchers introduce “Learning to Yield and Request Control (YRC),” including an open-source benchmark, a validation approach, and comparisons of learning methods across diverse environments.
Computational Frameworks for Human CareCHAI affiliate Brian Christian publishes an article in Daedalus on how AI alignment has progressed toward “care-like relationships,” co-edited under “The Social Science of Caregiving.”
A Practical Definition of Political Neutrality for AICHAI senior scientist Jonathan Stray proposes a practical definition of AI neutrality (“maximum equal approval”) and describes related evaluation ideas and concerns about political influence.
Getting By Goal Misgeneralization With a Little Help From a MentorA CHAI update on a paper (presented at a NeurIPS 2024 workshop) studying whether allowing an agent to ask for help can mitigate goal misgeneralization under distribution shift.
Linear Probe Penalties Reduce LLM SycophancyA CHAI update on a paper published at the NeurIPS SoLaR workshop describing a methodology to reduce sycophancy-like unwanted LLM behaviors.