AI Safety Research Program 2019

This fall we are organizing the AI Safety Research Program for students and researchers interested in the field of AI alignment.

The program will be slightly different from the previous AI safety camps: we will focus much more on the phase of problem selection and refinement, as well as better orientation in the AI alignment area and looking for impactful directions. This will in large part happen at a extra retreat early in the program; we will also move towards more ambitious and longer-term projects with expert mentoring.

Program overview

In the program, we bring together a great group of participants and find research projects to work on in teams. The program consists both of focused work time and structured content on research skills, mapping AI alignment, careers, mentoring, discussions, and researcher feedback. The goal is to work on interesting projects and gain research skills and knowledge of the AI alignment area.

The research program will consist of two research retreats and online collaboration periods between the retreats and after the program, overall spanning ~3 months. Most projects continue after the end of the program, some to finalize and publish their work, some become ongoing long-term projects.

The participants of AI safety camp 2 in Czech Republic.

Who should apply?

The intended participants are mostly students and early career researchers wanting to work on AI safety, people switching careers, or coming from unusual backgrounds. Participants should have some familiarity with one or more fields relevant to technical AI safety or strategy research. Note that the program capacity is limited.

Timeline

30 Sept 2019                  Application deadline (23:59 GMT)
20 Oct 2019                    Admission results (latest)
16 Nov – 20 Nov 2019    Topic selection workshop, Luckley Farm, United Kingdom
21 Nov – 07 Feb 2020     Online collaboration phase
08 Feb – 16 Feb 2010     Research retreat, Ecocentre Louti, Czech Republic (tentative date)
after 17 Feb                    Finalizing, publishing or further work on projects

Participants must take part in all parts of the program.

Funding

The AI Safety research program is funded by the Long Term Future Fund. The entire program is free for the participants. We may be able to provide financial support for travel costs, especially if you would not be able to attend otherwise.

Ready to apply?

Applications are open HERE. The deadline is 30th September 2019 (23:59 GMT). (We encourage you to apply earlier at your convenience; it will help us with timely processing.)

If you can’t make it this time but want to stay in the loop, subscribe to our mailing list here.

More questions?

Check out the FAQ section of this website. If in doubt, contact us at contact@aisafetycamp.com.

Research sessions at AISC 2.