AI Smith Lab builds tools and systems that make agentic AI safer to build and safer to use.

Working at the intersection of human-computer interaction, privacy and security, and systems, we design practical methods that help developers construct trustworthy agentic systems and help users interact with them safely and confidently.

The name of the lab comes from The Computer Scientist as Toolsmith II.
"A toolmaker succeeds as, and only as, the users of his tool succeed with his aid."