AI Security & Safety - Summer Research Bootcamp

CALL FOR PARTICIPANTS AI Security & Safety Summer Research Bootcamp WMG, University of Warwick · Summer 2026 · 8 Weeks (13th July - 4th September)

Overview

As artificial intelligence systems become increasingly integrated into critical infrastructure, society, and governance, ensuring their security, robustness, and alignment with human values is a global priority. The University of Warwick is running an intensive, 8-week Summer Bootcamp dedicated to AI Security and Safety. This program is uniquely designed to convene top-tier researchers from academia, industry, and government sectors to foster cross-disciplinary collaboration, accelerate foundational research, and establish best practices in the field.

The University of Warwick is pleased to invite researchers from academia, industry, and government to participate in an intensive 8-week Summer Research Bootcamp in AI Security and Safety. As artificial intelligence systems become increasingly embedded in critical infrastructure, healthcare, finance, and national security, the imperative to develop rigorous, research-led approaches to AI safety and adversarial robustness has never been greater.

This bootcamp brings together a diverse cohort of researchers to collaborate on cutting-edge problems at the intersection of machine learning, cybersecurity, and responsible AI governance. Participants will leave with tangible research outputs, expanded networks, and deepened expertise in one of the most consequential fields of our time.

Research Focus Areas

Security & Adversarial Robustness

• Adversarial Machine Learning
• Red-teaming and jailbreak mitigation for large language model
• Backdoor attacks, data poisoning, and supply-chain threats
• Privacy-preserving machine learning and federated systems

Safety, Alignment & Governance

• AI alignment and value learning techniques
• Interpretability, transparency, and explainability
• Risk assessment frameworks for autonomous systems
• Policy, regulation, and responsible deployment standards

Programme Structure

Weeks 1–2: Foundations & Onboarding

• Intensive orientation seminars in AI security and safety
• Project scoping and team formation workshops

Weeks 3–7: Active Research Phase

• Dedicated project work with faculty supervision
• Weekly expert seminars and invited scientists/practitioners talks
• Peer-review sessions and mid-programme presentations
• Technical training workshops on specialist tools

Weeks 8: Reporting

• Final research presentations and project showcase
• Publication and knowledge-transfer planning

Who Should Apply

Fees & Scholarship Opportunities

Participation Fee: £3,000

The participation fee covers access to all seminars, training workshops, research facilities, and bootcamp materials for the full 8-week programme. Accommodation and travel are not included for standard participants.

Early Career Researcher Scholarships

A limited number of fully-funded scholarships are available for exceptional students and early career researchers (within 5 years of PhD, or equivalent), covering:

• Full programme fee waiver (£3,000)
• On-campus accommodation for the 8-week duration
• Priority access

Partially-funded scholarships may also be available.

Scholarship applications require a short statement of motivation and CV. Places are limited — early application is strongly encouraged.

How to Apply

Applications are submitted via the University of Warwick online portal. Applicants should provide a current CV, a brief statement of research interest (up to 500 words), and, for scholarship applicants, a short motivation letter.

Application Deadline 13:00 BST, 22nd June

Notification of acceptance and scholarship (where applicable) will be given by 25th June.

Apply online at AI Security & Safety Summer Research Bootcamp – Fill out formLink opens in a new window

Enquiries: For any informal enquiries please contact Prof. Maple at cm@warwick.ac.uk

Selection criteria:

Candidates from both academia and industry will be assessed against the following criteria:

1. Relevance of background - the applicant's research interests, professional work, or academic training aligns meaningfully with one or more of the bootcamp's focus areas: adversarial robustness, AI security, AI safety and alignment, interpretability, or AI governance.
2. Demonstrated expertise - evidence of prior engagement with AI, machine learning, cybersecurity, or related fields, whether through publications, technical projects, professional roles, or academic study.
3. Capacity to contribute - the applicant can credibly contribute to a collaborative research environment, either through technical skills, domain knowledge, or complementary expertise (e.g. law, policy, social sciences).
4. Clarity of purpose - the statement of research interest articulates a clear and specific reason for attending, with plausible research questions or problems the applicant wishes to explore during the programme.
5. Career stage and institutional diversity - we aim to build a cohort fostering cross-sector collaboration with a balance of career stages (student, early career, senior) and a range of institutional types (universities, technology companies, public sector bodies, NGOs), to foster cross-sector collaboration.
6. Availability and commitment - the applicant is available for the full 8-week programme duration and can demonstrate institutional or employer support where relevant.

In addition to the acceptance criteria above, scholarship applicants will be assessed on:

1. Financial need or access barriers - the applicant demonstrates that the participation fee or accommodation costs would represent a genuine barrier to attendance without funding support.
2. Quality of motivation letter - the motivation statement goes beyond the research interest statement to articulate how the bootcamp will have a meaningful impact on the applicant's career development, research agenda, or the organisations and communities they serve.
3. Research potential and trajectory - evidence of research potential(e.g. technical reports, preprints, independent projects, patents, policy contributions, or strong academic references).
4. Strategic alignment - the applicant's stated aims align closely with the bootcamp's goal of producing tangible, publishable research outputs and contributing to best-practice development in responsible AI.
5. Underrepresentation - applications are particularly welcomed from individuals from groups that are underrepresented in AI security and safety research.