George Chalhoub

George Chalhoub

George Chalhoub studies the security, privacy, and safety of AI systems, from large language models to autonomous agents. His research brings together human-computer interaction, cybersecurity, and law to understand how people trust and delegate to AI, where these systems fail, and which safeguards make them safer and more accountable.

He holds a DPhil in Cyber Security from the University of Oxford (Pembroke College), where he was also a postdoctoral researcher, served as a Fellow at Harvard's Berkman Klein Center before becoming a Faculty Associate, and previously held research roles at Microsoft Research and Nokia Bell Labs in Cambridge. Endorsed by UK Research and Innovation as a Global Talent, his work appears at ACM CHI, ACM CSCW, USENIX SOUPS, ACM IMWUT, and IJHCS, holds a granted US patent with Microsoft, and has been featured in Fortune, Times Higher Education, Yahoo News, and more than thirty outlets worldwide.

Focus
  • AI agents. Security and oversight when autonomous agents browse the web, hold credentials, and coordinate with one another.
  • AI browsers and prompt injection. How AI-powered browsers blur data and instructions, and how design can bound failure.
  • Privacy in LLMs. Prompt disclosure and unintended data exposure across healthcare, law, journalism, and enterprise workflows.
  • AI governance. Regulation, platform policy, and accountable oversight, including norms for AI in scientific writing and peer review.
  • Human-centred security. How people understand and trust connected and AI systems, and how design keeps them safe.
Selected

All publications →

Links