RSS

Sahar Abdelnabi

Karma: 14

I am an AI security researcher at Microsoft. Previously, I was a PhD student at the CISPA Helmholtz Center for Information Security. I obtained my MSc degree at Saarland University, Germany. I am broadly interested in all things AI, security, safety, and evals.

Feel free to contact me at: sahar.s.abdelnabi@gmail.com and/​or saabdelnabi@microsoft.com

Do LLMs Com­ply Differ­ently Dur­ing Tests? Is This a Hid­den Vari­able in Safety Eval­u­a­tion? And Can We Steer That?

Sahar Abdelnabi16 Jun 2025 13:52 UTC
15 points
0 comments6 min readLW link