Author

M. Sriram

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Oct 2026

Do AI-for-Science Agents Preserve Model Safety?

This work asks whether model safety is preserved when an LLM is integrated in an AI-for-Science agent harness, and evaluates five agent harnesses on 50 harmful chemistry requests to show that the safety behavior of the base LLM is not durable once a harness is introduced.

Michael Lee, M. Sriram, N. Benjamin Erichson · 0 citations