Scholar
Andrew Paverd
Google Scholar ID: QZ2HOkMAAAAJ
Microsoft
Citations & Impact
Citations
2,388
H-index
23
i10-index
35
Publications
20
Co-authors
108
list available
Publications
Resume (English only)
Academic Achievements
- December 2024: Paper 'Get my drift? Catching LLM Task Drift with Activation Deltas' accepted for IEEE SaTML 2025.
- May 2024: Paper 'Trustworthy AI using Confidential Federated Learning: Federated learning and confidential computing are not competing technologies' published in ACM Queue.
- February 2024: Paper 'Closed-Form Bounds for DP-SGD against Record-level Inference Attacks' accepted for USENIX Security 2024.
- August 2023: Published the new 'Microsoft Vulnerability Severity Classification for AI Systems'.
- June 2023: Paper 'An Empirical Study & Evaluation of Modern CAPTCHAs' accepted for USENIX Security 2023.
- May 2023: Paper 'Bayesian Estimation of Differential Privacy' accepted for ICML 2023.
- April 2023: Paper 'SoK: Let The Privacy Games Begin! A Unified Treatment of Data Inference Privacy in Machine Learning' accepted for IEEE Symposium on Security and Privacy 2023.
- January 2023: Paper 'VICEROY: GDPR-/CCPA-compliant Enforcement of Verifiable Accountless Consumer Requests' accepted for NDSS 2023.
- May 2022: Published a paper on pre-hijacked accounts at USENIX Security 2022.
Research Experience
- April 2025: Will be giving lectures on AI Security at the CISPA-Ellis Summer School and KU Leuven Summer School.
- January 2025: Will give a talk on 'Vulnerability Response in the Era of AI' at EPFL AMLD 2025.
- August 2024: Will be giving a lecture at KU Leuven's Summer School.
- July 2024: Will be speaking on a panel about Generative AI and Security at the Centre for Digital Trust and Society Forum 2024.
- June 2024: Participated in the inaugural AI security tabletop exercise conducted by CISA.
- June 2023: Gave a keynote on Evaluating Privacy in Machine Learning at the Dependable and Secure Machine Learning (DSML) workshop 2023.
- January 2023: Opened applications for research internships in security and privacy at Microsoft Cambridge.
- May 2022: Published a paper on user account creation security failures at USENIX Security 2022.
- December 2021: Worked with Will Pearce and others.
Background
- Principal Research Manager at Microsoft Security Response Center (MSRC), focusing on AI security.
