Scholar
Aidan Ewart
Google Scholar ID: QjcaY50AAAAJ
Maths Undergrad @ University of Bristol
ML Safety
AI Safety
AI Alignment
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
854
H-index
5
i10-index
5
Publications
5
Co-authors
15
list available
Contact
Email
aidanprattewart@gmail.com
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
LinkedIn
Open ↗
Publications
3 items
AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviors
2026
Cited
0
Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities
2025
Cited
0
Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs
2024
Cited
72
Resume (English only)
Research Experience
Interned at Haize Labs and was a Research Scholar at MATS, mainly focusing on adversarial robustness.
Education
Studying at the University of Bristol, majoring in mathematics.
Background
An undergrad studying maths at the University of Bristol, working on making AI systems safe in free time.
Co-authors
15 total
Co-author 1
Lee D Sharkey
Goodfire
Aengus Lynch
University College London
Abhay Sheshadri
Undergraduate, Georgia Institute of Technology
Dylan Hadfield-Menell
Massachusetts Institute of Technology
Stephen Casper
PhD student, MIT
Henry Sleight
Research Manager, Anthropic Fellows Program, Program Manager, Constellation
Co-author 8
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up