AI Security Research Intern

The Microsoft Security Response Center (MSRC) works to protect customers and Microsoft from current and emerging threats to security and privacy. Within MSRC, our AI Vulnerabilities and Mitigations team analyzes all reported security vulnerabilities in Microsoft’s AI and agentic systems and develops new mitigations through deep research. 

Some of our team’s recent research and collaborations include:

• The Nine Dimensions of Prompt Injection Attacks: https://arxiv.org/pdf/2611.04672

• Securing AI Agents with Information-Flow Control (NeurIPS’26): https://arxiv.org/abs/2505.23643

• VerifiableFL: Verifiable Claims for Federated Learning using Exclaves (SoCC’26): https://arxiv.org/abs/2412.10537

• LLMail-Inject: A Dataset from a Realistic Adaptive Prompt Injection Challenge (EMNLP’26): https://arxiv.org/abs/2506.09956

• Optimizing Agent Planning for Security and Autonomy (ICLR’26) https://openreview.net/forum?id=g0aVCDY3gS

• Implicit Memory as a Hidden Channel in LLMs (SaTML’26): https://arxiv.org/pdf/2602.08563

• Highlight & Summarize: RAG without the Jailbreaks: https://arxiv.org/abs/2508.02872

• The Hawthorne effect in reasoning models: Evaluating and steering test awareness (NeurIPS’25): https://arxiv.org/abs/2505.14617

• TaskTracker: Catching LLM Task Drift with Activation Deltas (SaTML’25): https://arxiv.org/abs/2406.00799

We are looking for research interns to work with us on projects such as: identifying new classes of vulnerabilities in AI or agentic AI systems; detecting vulnerabilities in real systems; automating security vulnerability response using agents; and developing new mitigations that can be deployed at scale. 

Our team is uniquely placed to solve real-world security and privacy challenges, through cutting-edge scientific research, informed by vulnerability data from production systems.

• Contract Length: 12 weeks 
• Location: Cambridge, UK  

• Conduct cutting-edge research in AI security and privacy by proposing, exploring, and evaluating new ideas.
• Develop and implement prototypes to validate concepts and demonstrate real-world applicability.
• Collaborate with cross-functional teams to integrate security and privacy solutions into products and services.
• Publish findings and present insights to internal stakeholders and external conferences to advance industry knowledge.

Required/Minimum Qualifications:

• Students enrolled in a PhD program or outstanding undergraduate/master’s students with research experience (e.g., published research papers).

Preferred/Additional Qualifications:

• One or more papers at top security conferences (e.g., USENIX, CCS, S&P, NDSS) or papers focusing on security, safety, or privacy at top machine learning conferences (e.g., NeurIPS, ICML, ICLR).