Applications are now open for the Winter 2023-24 cohort of MATS (previously SERI MATS). Our mentors are: Adrià Garriga Alonso, Alex Turner, Buck Shlegeris, Caspar Oesterheld, Daniel Murfet, David ‘davidad’ Dalrymple, Erik Jenner, Ethan Perez, Evan Hubinger, Francis Rhys Ward, Jeffrey Ladish, Jesse Clifton, Jesse Hoogland, Lee Sharkey,  Neel Nanda, Owain Evans, Stephen Casper, Vanessa Kosoy, and researchers at Sam Bowman's NYU Alignment Research Group, including Asa Cooper Stickland, Julian Michael, Shi Feng, and David Rein.

Submissions for most mentors are due on November 17 (and for Neel Nanda on November 10). Many mentors ask challenging candidate selection questions, so make sure you allow adequate time to complete your application. We encourage prospective applicants to fill out our brief interest form to receive program updates and application deadline reminders. You can also fill out our recommendation form to let us know about someone who might be a good fit, and we will share our application with them.

We are currently funding constrained and accepting donations to support further research scholars. If you would like to support our work, you can donate here!

Program Details

MATS is an educational seminar and independent research program (40 h/week) in Berkeley, CA that aims to provide talented scholars with talks, workshops, and research mentorship in the field of AI alignment, and connect them with the Berkeley AI safety research community. MATS provides scholars with housing in Berkeley, CA, as well as travel support, a co-working space, and a community of peers. The main goal of MATS is to help scholars develop as AI safety researchers. You can read more about our theory of change here.

Based on individual circumstances, we may be willing to alter the time commitment of the program and arrange for scholars to leave or start early. Please tell us your availability when applying. Our tentative timeline for the MATS Winter 2023-24 program is below.

Scholars will receive a USD 12k stipend from AI Safety Support for completing the Training and Research Phases.

Applications (Now!)

Applications open: October 20

Applications are due: November 17

Note: Neel Nanda's applicants will follow a modified schedule; see section below.

Training Phase (January 8 - January 21)

Accepted applicants are generally expected to have completed the AI Safety Fundamentals Alignment Course or similar prior to the start of the MATS program.

MATS begins with a two-week Training Phase. To equip scholars with a broad understanding of the AI safety field, the Training Phase features an advanced AI safety research curriculum, mentor-specific reading lists, discussion groups, and more.

Research Phase (January 22-March 15)

The core of MATS is a two-month Research Phase. During this Phase, each scholar spends at least one hour a week working with their mentor, with more frequent communication via Slack. Mentors vary considerably in terms of their:

  • Influence on project choices;
  • Attention to low-level details vs high-level strategies;
  • Emphasis on outputs vs processes;
  • Availability for meetings.

Our Scholar Support team complements mentors by offering dedicated 1-1 check-ins, research coaching, debugging, and general executive help to unblock research progress and accelerate researcher development.

Educational seminars and workshops will be held 2-3 times per week. We also organize multiple networking events to acquaint scholars with researchers in the Berkeley AI safety community.

Research Milestones

Scholars complete two milestones during the Research Phase. The first is a Scholar Research Plan outlining a threat model or risk factor, a theory of change, and a plan for their research. This document will guide their work during the remainder of the program, which culminates in a research symposium attended by members of the Berkeley AI safety community. The second milestone is a ten-minute research presentation at this event.

Community at MATS

The Research Phase provides scholars with a community of peers, who share an office, meals, and housing. In contrast to pursuing independent research remotely, working in a community grants scholars easy access to future collaborators, a deeper understanding of other research agendas, and a social network in the AI safety community. Scholars also receive support from a full-time Community Manager.

During our Summer Cohort, each week of the Research Phase included at least one social event, such as a party, game night, movie night, or hike. Weekly lightning talks provided scholars with an opportunity to share their research interests in an informal, low-stakes setting. Outside of work, scholars organized social activities, including road trips to Yosemite, visits to San Francisco, pub outings, weekend meals, and even a skydiving trip.

Extension Phase (April 1-July 26)

At the conclusion of the Research Phase, scholars can apply to continue their research in a four-month Extension Phase cohort, in London by default. Acceptance decisions are largely based on receiving mentor endorsements and securing external funding. By this Phase, we expect scholars to pursue their research with high autonomy.

Post-MATS

Since its inception in late 2021, the MATS program has supported 155 scholars and 26 mentors. After completion of the program, MATS alumni have:

You can read more about MATS alumni here.

Information Specific to Neel Nanda

Applications

Applications are due: November 10

Acceptance decisions: November 17

Training Phase (November 20 - December 22)

Neel Nanda's scholars will complete a four-week remote Training Phase, consisting of two weeks learning about mech interp and two weeks performing a Research Sprint in a pair—see the overview doc from the last program for more information.

Significantly more offers will be made for this initial Training Phase than for the subsequent Research Phase (last time, 22 individuals completed this Phase, and nine continued to the Research Phase), but past scholars who did not progress still found this Phase a good introduction to mech interp research. Offers to continue will be based largely on performance in the Research Sprint.

Neel's trainees will receive a stipend of USD 4800 from AI Safety Support for participation in the remote Training Phase.

Research Phase (January 8 - March 15)

Those who continue to the Research Phase will begin in-person in Berkeley on January 8 with the other scholars. For Neel's scholars, this will be considered the start of the Research Phase.

Who Should Apply?

Our ideal applicant has:

  • An understanding of the AI safety research landscape equivalent to having completed AI Safety Fundamentals' Alignment Course (if you are accepted into the program but have not previously completed this course, you are expected to do so before the Training Phase begins);
  • Previous experience with technical research (e.g. ML, CS, math, physics, neuroscience, etc.), generally at a postgraduate level; and
  • Strong motivation to pursue a career in AI safety research.

Even if you do not meet all of these criteria, we encourage you to apply! Several past scholars applied without strong expectations and were accepted.

Applying from Outside the US

Scholars from outside the US can apply for B-1 visas (further information here) for the Research Phase. Scholars from Visa Waiver Program (VWP) Designated Countries can instead apply to the VWP via the Electronic System for Travel Authorization (ESTA), which is processed in three days. Scholars who receive a B-1 visa can stay up to 180 days in the US, while scholars accepted into the VWP can stay up to 90 days. Please note that B-1 visa approval times can be significantly longer than ESTA approval times, depending on your country of origin.

How to Apply

Applications are now open. Submissions for most mentors are due on November 17. We encourage prospective applicants to fill out our brief interest form to receive program updates and application deadline reminders. You can also fill out our recommendation form to let us know about someone who might be a good fit, and we will share our application with them.

Candidates apply to work under a particular mentor, who will review their application. Applications are evaluated primarily based on responses to mentor questions and prior relevant research experience. Information about our mentors' research agendas and application questions can be found on the MATS website.

Before applying, you should:

  • Read through the descriptions and agendas of each stream and the associated candidate selection questions;
  • Prepare your answers to the questions for streams you’re interested in applying to. These questions can be found on the application;
  • Prepare your LinkedIn or resume.

The candidate selection questions can be quite hard, depending on the mentor! Make sure you allow adequate time to complete your application. A strong application to one mentor may be of higher value than moderate applications to several mentors (though each application will be assessed independently).

Note that the application is longer than it first seems because mentor-specific questions are hidden until you select a mentor.

Application Office Hours

We have office hours for prospective applicants to clarify questions about the MATS program application process. These sessions will begin with a brief presentation about the program. Before attending office hours, we request that applicants read the current post fully and our FAQ.

Our office hours will be held on this Zoom link at the following times:

  • Wednesday, October 25, 12 pm-2 pm PT.
  • Wednesday, November 1, 10 am-12 pm PT.
  • Wednesday, November 1, 4 pm-6 pm PT.

You can add these office hours to Google Calendar with this link.

New to LessWrong?

New Comment
6 comments, sorted by Click to highlight new comments since: Today at 9:56 AM

Does the financially support include housing (separate from the stipend)?

Interesting to see how last year's program paid 16k for 7 weeks of full-time research and this year's program pays 12k for 9 weeks of full-time training+research. (41% less.) Funding hits

Our alumni were surveyed about what amount of funding they would consider sufficient to participate in MATS and the median (discounting the $0 responses) was quite low.

Also, last year's program was 8-weeks and this year's program is 10-weeks.

Update: We have finalized our selection of mentors.