If none of these positions are the right fit for you but you would still like to express an interest in working with us, please fill out this form .
General enquiries about jobs at CHAI can be sent to chai-info@berkeley.edu .
Start Date: Flexible Compensation: Negotiated Apply: Via this form
CHAI Research Fellows will work with CHAI faculty members Stuart Russell, Pieter Abbeel, and Anca Dragan and will also have the opportunity to collaborate with CHAI affiliate faculty at Berkeley, other faculty members in Berkeley AI Research (where CHAI is located), and many other institutions. The Fellowship is aimed at training highly qualified postdoctoral researchers to carry out research to advance beneficial AI.
Duties of position
Given the broad and open-ended mandate of the Center to develop provably beneficial AI systems, the post holder will have considerable freedom to pursue novel research projects in areas such as reasoning, decision making, learning, multi-agent systems, and philosophical foundations using the tools of probability theory, game theory, control theory, etc. Topics of particular current interest include
Applications are handled by the employer or external application platform. AI Safety Careers does not process applications directly.
Theoretical foundations of intelligent agent architecture
Expressive formal languages for probability models, including probabilistic programming
Decision making over long time scales
Cumulative lifelong learning
Human-in-the-loop planning and reinforcement learning
Human-robot interaction
The structure and properties of human preferences
Foundations for acting on behalf of multiple humans
Robust cooperation in heterogeneous multiagent systems
Mechanism design for human-machine systems
Depending on their interests, Fellows may also lead research collaborations with other CHAI researchers and Berkeley PhD students, advise more junior student researchers, and teach courses relevant to their research. Candidates for the Fellowship are assessed based on their academic and research achievements, as well as on personal merit and leadership qualities.
Qualifications
Candidates must have (or be about to obtain) a PhD in a relevant technical discipline (computer science, statistics, mathematics, or theoretical economics) and a record of high-quality published research. A solid understanding of current methods in AI and statistical learning would be an advantage. Candidates need not have done previous work on the AI control problem.
Salary & Visa Sponsorship
The CHAI Research Fellowships provide funding for two years, renewable up to three years given satisfactory performance. Salary will be commensurate with qualifications and experience and based on UC Berkeley salary scales. The position is eligible for visa sponsorship. CHAI will also contribute to visa costs for Fellows and their dependents (partner and children).
Application
Apply via this form . Supporting documents include a resume, a one-page statement of interest describing in general terms the kind of research you would like to undertake at CHAI, and the names and email addresses of two academic referees.
Our core research spans computer science, mathematics, control theory, robotics, statistics, formal logic, economics (including game theory), cognitive psychology, or neuroscience. We may also be interested in moral philosophy, sociology, political science, law, and other fields dealing with formal and semiformal theories of human value systems.
If you are interested in collaborating on research that aligns with our mission , please fill out this form .
NOTE: Undergraduates interested in graduate study at Berkeley in this area should apply directly to the appropriate UC Berkeley graduate program. You may mention your interest in CHAI on your application, but we are unable to review or give feedback on your application. We will be happy to discuss research and/or supervision once you have been admitted.
Applications for 2026 internships are now closed.
This internship is designed for individuals who are interested in research in human-compatible AI. Our internships require a background in machine learning, computer science, mathematics, or related fields. Existing research experience in machine learning is strongly advantageous but not required. We are interested in people who can demonstrate technical excellence and wish to transition to research in human-compatible artificial intelligence. Applicants may include undergraduate or Master’s students in computer science or adjacent fields, PhD students/researchers, professional software or ML engineers, etc.
CHAI interns work on a research project supervised by—and often proposed by—a mentor. Most projects are primarily executed by the intern, while the mentor provides advice and guidance; these are typically published as first-author workshop papers. On occasion, a mentor may propose a more team-based project that they or other interns are also executing on; these are typically published as conference papers. In addition, interns are encouraged to participate in other activities in CHAI, such as our weekly seminars, annual workshop and team socials.
All applicants should aim to understand CHAI’s research interests before applying. See our mentor profile page for a list of mentors and their interests.
General Information
Location: In-person (at UC Berkeley) is preferred; remote is possible.
Start Date: Flexible. We typically hold internships over the summer, but dates are flexible throughout the year and you’ll coordinate with your mentor.
Duration: Internships are typically 12 to 16 weeks.
Compensation: $5,000 per month for in-person interns and $3,500 per month for remote interns.
International Applicants: We accept international applicants.
Internship Application Process Overview
The internship application process has four phases. Please note: while we will do our best to adhere to these dates, they are subject to change.
Initial Review (Phase 1) We will examine your application based on motivation, research potential, grades, experience, programming ability, and other criteria. We will begin reviewing applications after the application closes. Applicants will receive a response by mid October.
We will examine your application based on motivation, research potential, grades, experience, programming ability, and other criteria.
We will begin reviewing applications after the application closes.
Applicants will receive a response by mid October.
Programming Assessment (Phase 2) You will be given a programming test. Applicants will receive a response by early December.
You will be given a programming test.
Applicants will receive a response by early December.
Interview Rounds (Phase 3) We typically conduct two rounds of interviews which will occur between late November and mid December. We have several mentors who are willing to take on interns. Each mentor that is interested in working with you will contact you to schedule an interview. It’s possible that you will speak to multiple mentors.
We typically conduct two rounds of interviews which will occur between late November and mid December.
We have several mentors who are willing to take on interns. Each mentor that is interested in working with you will contact you to schedule an interview. It’s possible that you will speak to multiple mentors.
Offer (Phase 4) Applicants will likely receive offers by mid December. If you are given an offer by one mentor, then you will work with that mentor if you choose to take the internship. If you are given multiple offers from different mentors, then you may choose the mentor you’d like to work with. While our internships tend to begin around May, the start date will ultimately depend on you and your mentor(s). You and your mentor(s) will coordinate when to begin your internship.
Applicants will likely receive offers by mid December.
If you are given an offer by one mentor, then you will work with that mentor if you choose to take the internship.
If you are given multiple offers from different mentors, then you may choose the mentor you’d like to work with.
While our internships tend to begin around May, the start date will ultimately depend on you and your mentor(s). You and your mentor(s) will coordinate when to begin your internship.
You will hear from chai-internship@berkeley.edu with either a rejection or an offer. If you have not heard, please exercise patience as we carefully review our candidates.
Contact
If you have any questions, please contact us at chai-internship@berkeley.edu .
Everyone is welcome at CHAI. To request accommodations at any point during the hiring process, you can email us at chai-internship@berkeley.edu . We won’t ask you to disclose the reason for your accommodation.
Additional Information
To view our former interns’ profiles, click here . Former interns have developed essential skills and knowledge, preparing them for graduate school and industry opportunities. Take a look at what some of them had to say about their internship experience:
Beth Barnes
I thoroughly enjoyed my time at CHAI, and learnt a lot. I came away with a broad understanding of the inverse RL field, much improved coding ability, and a sense of how to approach research. Dylan was a really dedicated supervisor, giving me lots of help with all the components of research, from choosing a question to fixing bugs! – Beth Barnes
Stephen Casper
Interning at CHAI was immensely valuable for building skills and connections involving AI safety research. My project involved developing and testing a set of interpretability techniques for neural networks, and having it under my belt is great experience and will be helpful in pursuing additional research work in the future. Meanwhile, the institutional culture in CHAI is both very friendly and conducive to a heavy flow of ideas through biweekly seminars and advising meetings. Overall, I think this would be a great opportunity for anyone who wants to start doing high-impact AI safety research. – Stephen Casper
Chris Cundy
I spent the summer working with some of the leading researchers in AI safety and completed a substantial research project, resulting in co-authorship on two academic papers submitted after I left. The chance to enjoy summer in California was an awesome bonus. – Chris Cundy
Dmitrii Krasheninnikov
The internship was a very enjoyable and intense learning experience. I acquired an in-depth understanding of inverse RL & utility aggregation, connections with the leading researchers in AI safety and a skill of noticing being even a tiny bit confused about the problem and zooming in on the confusion. I expect that all of these will be extremely helpful in my research career. – Dmitrii Krasheninnikov
Cynthia Chen
The internship at CHAI was a truly valuable and enjoyable experience. My work was on benchmarking representation learning algorithms for imitation learning, and I’ve gained a lot of insights into self-supervised learning throughout this project. The people at CHAI are very friendly and supportive, and I constantly find myself impressed by their capabilities and hard work. In general, for anyone passionate about conducting beneficial AI research and joining conversations with many brilliant minds, I think interning at CHAI will be an excellent opportunity to learn and grow to become a better researcher. – Cynthia Chen
Harry Giles
I spent the summer working on the Inverse Reinforcement Learning problem (IRL). This problem is interesting because in many domains – e.g. self driving cars – we have examples of expert behaviour, but specifying an accurate reward function is very hard. Moreover, the problem is difficult because of degeneracy; that is, there is often a large class of reward functions which explain a fixed optimal or noisy optimal policy. My project addressed the assumption of stationarity in IRL, whereby it is assumed that the expert policy does not change during demonstrations. This may be an unreasonable assumption in practice because an expert demonstrator might themselves be improving over time. My work proposed a simple model to account for the non-stationarity of demonstrations, and showed empirically that in some cases the task of reward inference can be made easier when working with non-stationarity, even overcoming some of the degeneracy problems from stationary IRL. I had a wonderful time and I am very happy with our results, which we will be presenting at a NeurIPS workshop later this year. – Harry Giles
Michael McDonald
Even with the challenges of the COVID-19 pandemic and having to conduct research remotely, my internship with CHAI was an incredibly beneficial experience. Thanks to the close interaction with my mentor, Dylan, and the abundance of resources provided to support me and my fellow interns, it really felt like I was part of an active and passionate community (even though none of us ever met in person). My project focused in robotics, with the core idea of using hierarchical imitation learning to train neural networks that emulate traditional task and motion planning systems. This will (hopefully) serve as an important bridge from slower but more reliable planning systems, to fickle and less predictable deep learning based approaches. For the entire course of the project, whenever I was at all uncertain over how to proceed or not quite asking the right questions, Dylan was there to help me not only course-correct but also develop the skills to better handle those situations myself in the future. All-in-all, it was an experience I would highly recommend to anyone interested in AI-safety and wanting to grow as a researcher. – Michael McDonald
Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game by Sam Toyer, Olivia Watkins, Ethan Adrian Mendes, Justin Svegliato, Luke Bailey , Tiffany Wang, Isaac Ong, Karim Elmaaroufi, Pieter Abbeel, Trevor Darrell, Alan Ritter, and Stuart Russell, 2023. (ICLR 2024 spotlight; NeurIPS 2023 workshop)
Mitigating Generative Agent Social Dilemmas by Julian Yocum , Phillip Christoffersen, Mehul Damani, Justin Svegliato, Dylan Hadfield-Menell, and Stuart Russell. (FMDM, NeurIPS‑23)
UC Berkeley Values and Principles
We are looking for equity-minded applicants who represent the full diversity of California and who demonstrate a sensitivity to and understanding of the diverse academic, socioeconomic, cultural, disability, gender identity, sexual orientation, and ethnic backgrounds present in our community. When you join the team at Berkeley, you can expect to be part of an inclusive, innovative and equity-focused community that approaches higher education as a matter of social justice that requires broad collaboration among faculty, staff, students and community partners. In deciding whether to apply for a position at Berkeley, you are strongly encouraged to consider whether your values align with our Guiding Values and Principles, our Principles of Community, and our Strategic Plan.
Equal Employment Opportunity
The University of California is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, or protected veteran status. For more information about your rights as an applicant see here . For the complete University of California’s Nondiscrimination Policy, see here .
Applications are handled by the employer.
AI Safety Careers does not process applications directly.
This listing may be aggregated from a public source or submitted by a third party. If you represent this employer and would like to update or remove this listing, contact support@aisafetycareers.com.