Can LLMs do security code review?
Invited presentation on using large language models in security code review: where they help, where they fail, and how to keep human judgment and evidence in the loop.
This deck is based on publicly disclosable ideas—published practice, open discussion of AI-assisted review, and general non-secret methodology. It does not disclose confidential assessments or private vulnerability details.
View slides
Use the arrows (or ← → / Space when the viewer is focused) to move slide by slide.
Download: PowerPoint (.pptx)
Related writing on this site: Run AI-assisted security code review, Keep AI review trustworthy, Security code review minibook, Security code review trends in the AI era.