Crypto professor builds an AI-assisted one-way function transform lab for interactive security proofs

jedisct1 · x · 2026-09-21

Cryptography researcher Chris Peikert built a "one-way function transform lab" with AI help for his class: an interactive site where you write security proofs (reductions + analysis) and attacks — both correct and deliberately wrong — for a variety of natural function transforms. A hands-on experiment blending LLM-assisted tooling with crypto pedagogy, now live on the web.

Original post →

More from coding & agent

coding & agent channel →