The MATS Research fellowship is a highly competitive, fully funded research and training program designed to prepare the next generation of researchers working on AI alignment, AI security, and frontier AI risk reduction. Hosted in major research hubs such as Berkeley and London, the program provides participants with financial support, mentorship from world-leading AI researchers, and access to substantial compute resources.
This fellowship is designed for individuals who want to contribute meaningfully to ensuring advanced artificial intelligence systems are safe, transparent, and beneficial to humanity.
Program Overview
The MATS Research fellowship is structured as an intensive 10 to 12-week in-person research program that brings together talented fellows and mentors from top AI institutions. Participants work directly with experts from organizations such as:
- Anthropic
- OpenAI
- Google DeepMind
- Redwood Research
- ARC and other frontier AI safety groups
The program is designed to provide a deeply immersive environment where fellows can focus entirely on research without financial or logistical barriers.
Key Program Details
- Location: Berkeley (USA) or London (UK)
- Duration: 10–12 weeks
- Start Date: September 28, 2026
- Application Deadline: June 7, 2026
- Funding Package Includes:
- $12,500 stipend
- $20,000 compute budget
- Free housing
- Meals provided
- Travel costs covered
- J-1 visa support (if required)
Research Tracks Available
Fellows can apply to one or multiple research tracks depending on their interests and skills:
- Empirical Research
- Theoretical Research
- Strategy and Forecasting
- Policy and Governance
- Systems Security
- Field-building and Founding
- Biosecurity
Each track is designed to address different dimensions of AI safety, from technical alignment research to governance and societal impact analysis.
Mentorship and Research Environment
One of the strongest aspects of the MATS Research is its mentorship model. Fellows are paired with experienced researchers from leading AI labs and safety organizations.
Mentors provide:
- Direct research guidance
- Project direction and feedback
- Technical and conceptual mentorship
- Support in publishing and scaling research ideas
This mentorship-driven structure allows fellows to rapidly grow their research skills while contributing to real-world AI safety work.
Fellowship Benefits
The program provides a comprehensive support system so fellows can focus fully on research:
Financial and Material Support
- $12,500 stipend
- $20,000 compute allocation
- Free accommodation
- Free meals
- Travel reimbursement
Academic and Professional Growth
- Collaboration with top AI researchers
- Access to cutting-edge AI safety problems
- Opportunities to publish research papers
- Exposure to frontier AI labs and institutions
Community and Networking
- Cohort-based learning environment
- Strong global AI safety community
- Networking with researchers and organizations in AI governance and technical alignment
Application Process
The application process is multi-stage and designed to evaluate both technical ability and research potential.
Stage 1: General Application
- Open until June 7, 2026
- Takes 1–2 hours
- Includes background questions and track-specific responses
- Requires contact details for two references
Stage 2: Central Evaluation
Applicants who pass Stage 1 proceed to standardized assessments:
- Empirical Track: research taste and technical implementation tests
- Policy & Governance Track: analytical writing evaluation
- Other tracks: may proceed directly to Stage 3
Stage 3: Stream Applications
- Applicants rank specific research streams
- Submit tailored responses or research proposals
- May include interviews or work tests
Final Offers
- Decisions finalized in late July / early August 2026
- Matching process considers both mentor and applicant preferences
- Program begins September 28, 2026
Impact and Outcomes
The MATS Research has built a strong track record of impact in the AI safety ecosystem:
- 80% of alumni work in AI alignment, transparency, or security fields
- 190+ research publications in recent years
- Over 12,000 citations across research outputs
- Strong presence in frontier AI organizations and academic institutions
Why This Fellowship Matters
As AI systems become increasingly powerful, the need for robust safety research grows rapidly. This program is designed to:
- Develop skilled researchers in AI alignment
- Support early-career talent in high-impact research
- Strengthen global AI safety infrastructure
- Bridge academia, industry, and policy communities
Participants often go on to work at leading AI labs or contribute to influential safety research initiatives.
Official Application Link
Learn more and apply here
Conclusion
The MATS Research fellowship is a rare opportunity for aspiring researchers to engage directly with frontier AI safety work in a highly supported environment. With full funding, mentorship from top institutions, and access to significant compute resources, it is one of the most impactful programs for those interested in shaping the future of safe artificial intelligence.
Click here for more global Fellowship opportunities
