The AI Alignment Foundation has opened applications for an eight-week, fully remote AI alignment research fellowship offering participants a $12,000 stipend to conduct research into artificial intelligence safety and alignment.
The fellowship is scheduled to run from September 8 to October 30, 2026, giving selected participants an opportunity to spend eight weeks working on AI safety research while receiving financial and technical support.
Applications for the programme close on August 17, 2026, making it a timely opportunity for researchers, technical professionals and other candidates interested in entering the rapidly developing field of AI safety.
AI Alignment Foundation Fellowship 2026
The AI Alignment Foundation’s fellowship is designed to support individuals interested in researching how increasingly capable artificial intelligence systems can be developed and deployed safely.
The eight-week programme is fully remote, allowing selected fellows to participate without relocating.
Successful participants will receive:
- $12,000 stipend for the eight-week fellowship.
- Payment of the stipend in two installments.
- Compute credits for research experiments.
- API credits to support research work.
- A fully remote research environment.
- An opportunity to complete and publish a substantial research write-up.
The provision of computing and API resources is particularly relevant for participants who may otherwise struggle to finance the infrastructure needed to conduct AI research.
Fellowship Dates
The 2026 AI alignment fellowship will run for approximately eight weeks.
Key dates include:
- Programme begins: September 8, 2026
- Programme ends: October 30, 2026
- Application deadline: August 17, 2026
- Duration: 8 weeks
- Format: Fully remote
- Stipend: $12,000
Applicants should ensure that they are available to participate throughout the fellowship period.
Who Should Apply?
The fellowship is intended for people who are interested in AI safety and alignment research and can demonstrate the ability to think carefully about challenging technical or conceptual problems.
Importantly, applicants do not need to have an existing AI alignment publication to apply.
Instead, the programme places emphasis on an applicant’s ability to:
- Think rigorously about difficult problems.
- Conduct meaningful research.
- Develop and investigate a research question.
- Produce a substantial written research output.
- Complete a project within the eight-week programme.
- Engage seriously with AI safety and alignment questions.
This makes the fellowship potentially suitable for talented individuals who are interested in entering AI safety research but have not yet built an extensive publication record in the field.
Research and Practical Experience
The programme is being run in collaboration with AE Studio’s alignment team.
The team has experience producing research in AI alignment and has worked on published research with major AI laboratories, including Anthropic.
Participants will therefore have an opportunity to engage with researchers working on practical AI alignment questions rather than simply undertaking a conventional academic course.
The fellowship’s short duration also means that fellows are expected to work efficiently and produce tangible research outputs during the eight-week period.
What Is AI Alignment?
AI alignment broadly refers to research focused on ensuring that artificial intelligence systems behave in ways that are consistent with intended human goals, values and safety requirements.
As AI systems become increasingly capable, researchers are investigating questions around reliability, robustness, interpretability, oversight and the potential risks associated with advanced AI systems.
AI safety has consequently developed into a growing interdisciplinary research area attracting computer scientists, mathematicians, engineers, social scientists and researchers from other backgrounds.
The fellowship provides an opportunity for participants to gain practical experience in this emerging field.
Financial and Research Support
One of the major features of the programme is its combination of financial support and research infrastructure.
Selected fellows will receive $12,000 over eight weeks, paid in two installments.
In addition, the programme provides compute and API credits.
This means fellows do not have to independently finance all of the computational resources required for their research experiments.
For applicants who have the skills and ideas to pursue AI safety research but lack access to funding or infrastructure, this support could significantly lower the barriers to participating in the field.
Expected Research Output
The programme is designed around producing a tangible research outcome.
Participants are expected to complete a research project and produce a write-up at the end of the eight-week fellowship.
Applicants should therefore be prepared to move from identifying a research question to investigating it and communicating their findings within a relatively short period.
The ability to execute a project and deliver a high-quality written output is likely to be an important consideration for prospective applicants.
How to Apply
Interested candidates should submit an application through the AI Alignment Foundation’s official fellowship application page.
Applicants should demonstrate their ability to reason carefully about difficult problems and explain their interest in AI safety research.
Previous AI alignment publications are not required, according to the programme information provided.
Candidates should nevertheless be prepared to demonstrate relevant skills, research ability, technical competence or other evidence that they can successfully complete an independent research project during the fellowship.
Why Consider the AI Alignment Fellowship?
The fellowship combines several features that can make it attractive to aspiring AI safety researchers:
- A $12,000 stipend.
- Eight weeks of focused research experience.
- Fully remote participation.
- Compute and API credits.
- Collaboration with an experienced alignment research team.
- An opportunity to produce a substantial research write-up.
- A potential entry point into AI alignment research without requiring a previous alignment publication.
For individuals who have considered entering AI safety research but have been limited by funding, infrastructure or lack of formal research experience, the programme could provide a structured opportunity to test their abilities and produce tangible work.
Application Deadline
The deadline to apply for the AI Alignment Foundation’s 2026 fellowship is:
August 17, 2026.
The programme is scheduled to begin on September 8, 2026, and conclude on October 30, 2026.
Because the application deadline is approaching, interested candidates should review the official fellowship information and submit their applications before the closing date.
AI Alignment Foundation Fellowship 2026 at a Glance
Programme: AI Alignment Foundation Research Fellowship
Field: AI Safety / AI Alignment Research
Duration: 8 weeks
Format: Fully remote
Dates: September 8 – October 30, 2026
Stipend: $12,000
Payment: Two installments
Research support: Compute and API credits
Previous AI alignment publication: Not required
Application deadline: August 17, 2026
Collaboration: AE Studio alignment team
