The Cambridge ERA:AI Fellowship 2027 is a fully funded, 10-week research fellowship in Cambridge, United Kingdom, designed for talented researchers, professionals, policymakers, and individuals from diverse academic and professional backgrounds who want to work on some of the most important challenges surrounding advanced artificial intelligence.

The Winter 2027 cohort will begin on 18 January 2027 and will provide selected fellows with financial support, expert mentorship, research assistance, networking opportunities, and access to a community working at the intersection of AI safety, AI governance, technical research, and public policy.

With a £10,000 stipend, visa support, meals during working hours, and travel expense coverage, the fellowship offers participants an opportunity to spend 10 weeks developing a substantial research project while working alongside experienced researchers and mentors in Cambridge.

Cambridge ERA:AI Fellowship 2027 at a Glance

• Opportunity: Cambridge ERA:AI Fellowship
• Cohort: Winter 2027
• Location: Cambridge, United Kingdom
• Duration: 10 weeks
• Start date: 18 January 2027
• Stipend: £10,000
• Funding: Fully funded
• Application deadline: 13 September 2026 at 11:59 p.m. Anywhere on Earth
• Minimum age: 18 years
• Application: Online
• Research areas: Technical AI Safety, AI Governance, and Technical AI Governance
• Mentorship: Yes
• Visa support: Yes
• Travel support: Yes
• Meals: Provided during working hours
• Potential continuation: 6+ month fully funded Research Scholars positions may be available to top fellows

What Is the Cambridge ERA:AI Fellowship?

The Cambridge ERA:AI Fellowship is a 10-week research programme hosted in Cambridge, UK. It is focused on helping talented individuals investigate ways of reducing risks associated with increasingly capable and autonomous AI systems.

The programme brings together people from different disciplines and professional backgrounds. Rather than limiting applications to individuals with a particular degree, nationality, profession, or academic credential, ERA takes a talent-first approach.

Participants are expected to develop a research project addressing important questions in frontier AI safety, AI governance, or technical mechanisms that can make AI governance more effective.

The fellowship is particularly relevant as artificial intelligence systems become increasingly capable and are being integrated into areas ranging from scientific research and cybersecurity to public administration, business, education, and national security.

What Does the Fellowship Provide?

One of the major attractions of the Cambridge ERA:AI Fellowship is that fellows are provided with extensive financial and professional support throughout the programme.

1. £10,000 Fellowship Stipend

Selected fellows receive a £10,000 stipend for the 10-week fellowship.

This financial support allows participants to dedicate substantial time to their research without having to treat the fellowship as an unpaid professional commitment.

2. Visa Support

For international fellows who require permission to participate in the programme in the United Kingdom, ERA provides visa support.

This is particularly important for applicants based outside the UK who want to participate in the in-person fellowship in Cambridge.

3. Travel Expense Coverage

The fellowship also provides support for travel expenses, helping reduce the financial barriers associated with travelling to Cambridge for the programme.

4. Meal Provisions

Fellows receive meal provisions during working hours, providing additional practical support during the 10-week programme.

5. Expert Mentorship

Each fellow works closely with a mentor during the fellowship.

Mentorship is an important component of the programme because participants are expected to develop and complete a meaningful research project. Mentors can provide guidance on research questions, methodology, project development, academic direction, and the broader implications of the fellow’s work.

ERA notes that mentors differ between cohorts. Fellows are matched with mentors after acceptance according to their research interests and needs.

Previous ERA mentors have included researchers and experts associated with organisations across academia, government, industry, and AI safety.

6. Research Management Support

Fellows receive dedicated research management support throughout their projects.

This support is intended to help participants become stronger researchers and policymakers in the AI safety and governance space.

The fellowship’s alumni have gone on to publish research in prominent academic venues, including conferences such as NeurIPS.

7. Cambridge Research Community

Participants become part of a community of people working on difficult questions related to advanced AI.

During the fellowship, fellows have access to a dedicated desk space at ERA’s Cambridge office and participate in regular social events and community activities.

This creates opportunities to exchange ideas, receive feedback, collaborate, and build professional relationships.

8. Networking and Career Development

ERA also helps fellows develop the skills, expertise, and networks necessary for careers in AI safety and AI policy.

The organisation can facilitate introductions to relevant organisations and professionals working in the field.

In exceptional cases, ERA may also provide additional financial assistance to support particularly impactful career transitions.

9. Potential Long-Term Research Opportunities

The fellowship can also serve as a pathway to longer-term research.

Top-performing fellows may have the opportunity to continue their work through ERA’s Research Scholars programme, which offers six months or longer of full-time funded positions.

This means that the 10-week fellowship can potentially become the beginning of a much longer research career in AI safety or governance.

Who Can Apply for the Cambridge ERA:AI Fellowship?

One of the most distinctive aspects of this fellowship is its broad eligibility policy.

Anyone aged 18 or older can apply.

ERA describes the programme as talent-first and states that it cares about much more than conventional credentials. There are no formal eligibility restrictions beyond being at least 18 years old.

Applicants therefore do not necessarily need to come from a traditional computer science or artificial intelligence background.

The programme welcomes people with different forms of expertise, provided they have the ability, motivation, and potential to contribute meaningfully to the mission of reducing risks associated with advanced AI.

Who Is Particularly Encouraged to Apply?

Although the fellowship is open broadly, it may be especially valuable for the following groups.

Researchers at Any Career Stage

Researchers can apply if they want to:

• Apply their existing expertise to high-priority AI safety or governance questions.
• Explore a new research direction.
• Work across disciplinary boundaries.
• Develop research relevant to advanced AI systems.
• Redirect an existing research programme toward frontier AI challenges.

The programme is not restricted to senior academics. Researchers at different stages of their careers may be considered.

Professionals from Technical and Policy Fields

Professionals working in areas such as:

• Technology
• Computer science
• Artificial intelligence
• Public policy
• Government
• National security
• Security studies
• Law
• Economics
• Research
• Governance
• Regulation
• Other relevant disciplines

may find the fellowship particularly valuable.

The programme provides an opportunity to apply existing professional expertise to emerging problems associated with increasingly capable AI systems.

People Redirecting Their Research Careers

The fellowship can also suit individuals who already have a research agenda but want to shift their work toward advanced AI safety and governance.

For example, someone with expertise in economics, cybersecurity, law, international relations, public administration, or computer science could use the fellowship to investigate how that discipline can contribute to managing risks from frontier AI.

Individuals with Distinctive Domain Expertise

ERA is also interested in applicants who bring unusual or underrepresented perspectives.

An applicant does not necessarily need to be an AI specialist if they possess expertise that could improve society’s understanding of advanced AI risks or contribute to developing effective solutions.

What Will Fellows Research?

During the fellowship, participants develop and complete a research project focused on technical and/or governance measures for mitigating risks posed by frontier AI systems.

Fellows receive support from:

• ERA research managers
• Individual mentors
• ERA’s wider institutional network
• Researchers and experts from relevant organisations

ERA supports research across the full range of frontier AI governance and technical AI safety.

Its wider research network includes experts associated with organisations such as the University of Cambridge, UK AI Security Institute, Centre for the Governance of AI, RAND Corporation, Google DeepMind, OpenAI, and Anthropic, among others.

The research programme is divided broadly into three streams.

1. Technical AI Safety

The technical stream focuses on understanding increasingly capable AI systems and developing methods to ensure that humans can maintain meaningful control and oversight.

As AI systems become more autonomous, researchers need to understand what these systems can do, how they behave, what capabilities they may develop, and how to prevent harmful or unpredictable outcomes.

ERA supports technical research in areas including:

• AI evaluations
• Interpretability
• Robustness
• Alignment
• AI control
• Dangerous capability detection
• Model behaviour monitoring
• Safeguards
• Human oversight
• AI system reliability

Research may involve investigating how to measure or elicit potentially dangerous capabilities, understanding the internal workings of AI models, designing stronger safeguards, monitoring model behaviour, and developing techniques for controlling systems whose objectives may not fully align with those of their operators.

The programme is also interested in emerging challenges such as:

• Detecting deceptive or strategically evasive AI behaviour.
• Controlling AI systems capable of automating parts of AI research and development.
• Accelerating alignment research across the LLM training stack.
• Developing safeguards that remain effective as AI models become increasingly capable.

The overall objective is to contribute to the development of AI systems that can remain subject to effective human oversight while incorporating appropriate safeguards against harmful or unpredictable behaviour.

2. AI Governance

The governance stream focuses on how governments, companies, international institutions, and other organisations should manage increasingly capable AI systems.

As frontier AI becomes more powerful, decisions regarding its development and deployment become increasingly consequential.

Effective governance requires institutions that can:

• Understand rapidly changing AI risks.
• Develop credible rules and safeguards.
• Hold powerful AI developers accountable.
• Coordinate across national borders.
• Respond to emerging technological developments.
• Build effective regulatory systems.

ERA supports research in areas such as:

• Regulatory design
• Regulatory implementation
• Frontier AI safety frameworks
• Corporate governance
• Accountability
• International coordination
• AI policy
• Institutional capacity

Researchers in this stream may investigate how governments and other institutions can develop governance systems that keep pace with technological progress.

Potential research questions may involve increasingly autonomous AI agents, international coordination, frontier AI regulation, and the development of security and verification systems for advanced AI.

The broader goal is to help institutions adapt as AI capabilities and associated risks change.

3. Technical AI Governance

Technical AI governance sits at the intersection of AI technology, governance, regulation, measurement, and verification.

The central idea is that effective AI governance cannot depend entirely on companies voluntarily reporting what their systems can do or how they were developed.

Governments and other stakeholders need reliable technical mechanisms for determining:

• What AI systems are capable of.
• How much computing power they use.
• Where systems are deployed.
• How systems are developed.
• Whether regulatory requirements are being followed.
• Whether claims made by AI developers can be independently verified.

ERA therefore supports research involving:

• Compute governance
• AI auditing
• AI monitoring
• Hardware-enabled verification
• Technical standards
• Measurement
• Privacy-preserving verification
• AI system auditing

Potential research could explore how regulators can identify AI systems that cross important risk thresholds, how hardware and computing infrastructure can enable verification while protecting privacy, and how increasingly autonomous systems can be monitored.

This field is particularly important because policymakers may create effective regulations but struggle to implement them if they lack the technical infrastructure needed to measure compliance.

ERA’s objective is to help develop governance systems based on credible measurement, verification, and enforcement rather than relying solely on trust or self-reporting.

Why Is the Cambridge ERA:AI Fellowship Important?

The rapid development of frontier AI has created a growing need for people who can understand both the opportunities and risks associated with increasingly capable systems.

AI safety is not solely a computer science problem. It involves questions concerning:

• Public policy
• Economics
• International relations
• Law
• Security
• Ethics
• Governance
• Computer science
• Engineering
• Institutional design
• Regulation

This makes interdisciplinary expertise increasingly valuable.

The Cambridge ERA:AI Fellowship provides an environment in which individuals from different backgrounds can bring their knowledge to emerging questions surrounding advanced AI.

For someone considering a career in AI safety, AI governance, technology policy, or related research, the fellowship can provide an opportunity to gain specialised experience while building connections with researchers and institutions working in the field.

Application Process for the Winter 2027 Fellowship

Applications for the Winter 2027 Cambridge ERA:AI Fellowship are currently open.

The selection process consists of several stages.

Stage 1: Written Application

The first stage primarily consists of short essay questions.

ERA estimates that applicants should spend approximately two hours completing this stage.

The written application is an important opportunity for candidates to demonstrate their thinking, motivation, research ability, and potential contribution to AI safety or governance.

Applicants should therefore approach the essays carefully rather than treating them as a simple administrative form.

Stage 2: First Interview

Following the written application stage, approximately the top 20% of candidates will be invited to first-round interviews.

These interviews are scheduled to take place in late September 2026.

Candidates should be prepared to discuss their interests, experience, reasoning, and potential research directions.

Stage 3: Second Interview

Applicants who are particularly promising after the first interview may be invited to a shorter second-round interview.

When candidates are invited to this stage, ERA will also contact their references.

Second-round interviews are expected to take place in early October 2026.

Stage 4: Offers and Research Development

Following the selection process, successful applicants will receive acceptance offers.

After offers are made, ERA will work with the selected fellows to develop their project ideas and match them with appropriate mentors.

This means applicants do not necessarily need to arrive with a completely finalised research proposal. The fellowship provides an environment in which project ideas can be refined with support from research managers and mentors.

Cambridge ERA:AI Fellowship Deadline

The application deadline for the Winter 2027 Fellowship is:

13 September 2026 at 11:59 p.m. Anywhere on Earth (AoE).

Applicants should avoid waiting until the final hours to submit their applications.

Because the deadline is expressed in Anywhere on Earth time, applicants should verify the corresponding deadline in their local time zone and submit well in advance.

How to Apply for the Cambridge ERA:AI Fellowship

Interested applicants can submit their application through the official application form.

Applicants should carefully review the application questions and provide clear, thoughtful responses demonstrating their suitability for the fellowship.

Tips for Preparing a Strong Application

Although the fellowship is talent-first and does not impose traditional academic eligibility requirements, applicants should still demonstrate why they are capable of making a meaningful contribution.

1. Clearly Explain Your Interest in Advanced AI

Show that you understand why AI safety and governance matter and explain what motivates you to work on these issues.

2. Connect Your Existing Expertise to AI

You do not necessarily need an AI degree.

Instead, explain how your current knowledge could contribute to AI safety, governance, or technical AI governance.

For example, a lawyer could explore AI regulation, an economist could investigate AI incentives and governance, a cybersecurity professional could examine AI security, while a political scientist could investigate international coordination.

3. Demonstrate Strong Research Thinking

The fellowship is fundamentally research-oriented.

Applicants should demonstrate an ability to identify important questions, think critically, analyse evidence, and develop useful approaches to complex problems.

4. Identify Potential Research Directions

While a fully developed proposal may not be required, having some ideas about the problems you would like to investigate can strengthen your application.

Think about questions that are:

• Important
• Relevant to frontier AI
• Researchable
• Potentially impactful
• Connected to your existing expertise

5. Highlight Your Distinctive Perspective

ERA specifically welcomes people with distinctive domain expertise and perspectives.

Applicants should therefore avoid assuming that they must compete by presenting themselves as conventional AI researchers.

Your unique professional or academic background may be one of your greatest strengths.

6. Take the Essay Questions Seriously

The written application is the first major stage of selection.

Because the programme expects applicants to spend around two hours on the essays, candidates should dedicate sufficient time to developing thoughtful and well-structured responses.

7. Prepare for Interviews

Candidates selected for interviews should be ready to explain:

• Why they are interested in AI safety or governance.
• What research questions interest them.
• How their existing expertise could contribute.
• How they approach difficult and uncertain problems.
• What they hope to accomplish during the fellowship.

Location and Fellowship Experience

The fellowship is hosted in Cambridge, United Kingdom.

Fellows will have dedicated desk space at the ERA office and will participate in community activities and social events.

The in-person environment provides opportunities to interact with other researchers and professionals while working on individual projects.

Cambridge also provides a strong academic and research environment, making it a particularly relevant location for individuals interested in advanced technology, public policy, and interdisciplinary research.

Who Should Consider Applying?

The Cambridge ERA:AI Fellowship may be particularly suitable for someone who:

• Wants to transition into AI safety research.
• Is already conducting research and wants to explore advanced AI questions.
• Works in technology or engineering.
• Has expertise in public policy or government.
• Works in law or regulation.
• Has a background in economics.
• Works in national security or cybersecurity.
• Wants to develop expertise in AI governance.
• Has interdisciplinary research interests.
• Possesses distinctive expertise that could contribute to understanding AI risks.
• Wants to build a professional network in AI safety and governance.
• Is interested in pursuing a longer-term research career.

Because the fellowship has no formal eligibility restrictions beyond the minimum age requirement, applicants should not automatically exclude themselves because they do not have a traditional AI background.

APPLY HERE

International Eligibility and Nationality

The fellowship states that anyone can apply, provided they are 18 years or older.

There is no stated restriction limiting applications to a particular race, nationality, country, or continent.

Therefore, applicants from Africa, Asia, Europe, North America, South America, Oceania, and other parts of the world may apply, subject to meeting the programme’s general requirements and being able to participate in the fellowship in Cambridge.

The fellowship’s broad approach means that applicants from countries across the Global South and Global North can potentially be considered.

International applicants should also take into account the practical requirements associated with travelling to and participating in a 10-week programme in the United Kingdom. ERA states that it provides visa support and travel expense coverage for fellows.

Visit HERE to learn more about the Cambridge ERA:AI Fellowship 2027 in the UK.

Key Reasons to Consider the Cambridge ERA:AI Fellowship

The fellowship combines several elements that can be difficult to access through conventional research opportunities:

1. A £10,000 stipend for the 10-week programme.
2. Visa support for eligible international participants.
3. Travel expense coverage.
4. Meals during working hours.
5. One-to-one mentorship.
6. Dedicated research management support.
7. Access to a Cambridge-based research community.
8. Networking opportunities across AI safety and governance.
9. Opportunities to develop research that can potentially lead to academic publication.
10. Potential access to 6+ month funded Research Scholars positions for top fellows.

The combination of financial support, mentorship, research experience, and professional networking makes the programme particularly relevant for individuals seeking to build a long-term career in AI safety or governance.

Conclusion

The Cambridge ERA:AI Fellowship 2027 offers an intensive 10-week opportunity for talented individuals to work on some of the world’s most important emerging questions surrounding advanced artificial intelligence.

The programme is particularly notable for its talent-first approach. Rather than restricting applications to a narrow academic discipline or professional background, ERA welcomes people with diverse expertise who can contribute to understanding and addressing frontier AI risks.

Successful fellows will receive a £10,000 stipend, visa support, travel expense coverage, meals during working hours, expert mentorship, research management assistance, workspace in Cambridge, networking opportunities, and access to a wider community working on AI safety and governance.

Research opportunities cover three major areas: Technical AI Safety, AI Governance, and Technical AI Governance. These streams address challenges ranging from AI alignment and interpretability to regulation, international coordination, auditing, monitoring, compute governance, and technical verification.

The programme is open to applicants aged 18 and above from any country or region, with no stated nationality, race, or continental restrictions. Applicants from technical, policy, legal, economic, security, research, and other relevant backgrounds are encouraged to consider applying.

For the Winter 2027 cohort, the fellowship begins on 18 January 2027, while applications close on 13 September 2026 at 11:59 p.m. Anywhere on Earth.

For candidates interested in building a research career focused on the safe development, governance, and responsible management of increasingly capable AI systems, the Cambridge ERA:AI Fellowship could provide a valuable platform to develop expertise, conduct meaningful research, establish professional connections, and potentially continue into longer-term funded research.