At the Lean FRO, Kim Morrison, a Senior Research Software Engineer, recently ran an experiment that went well beyond our expectations. An AI agent converted zlib, a widely used C compression library embedded in countless systems, to Lean, with minimal human guidance. No special tooling was built. It was Claude, a general-purpose AI, with no special training for theorem proving, out of the box. The workflow had four steps. First, the AI produced a clean, readable Lean implementation of the zlib compression format, including the DEFLATE algorithm at its core. Second, the Lean version passed the library’s existing test suite, confirming behavioral equivalence. Third, key properties were stated and proved, not as tests, but as mathematical theorems. The capstone theorem:
108 = /home/me/my_script.sh
，详情可参考一键获取谷歌浏览器下载
Мощный удар Израиля по Ирану попал на видео09:41
Tom walks through his first illegal trade in chillingly matter‑of‑fact terms: a contact gives him details of an upcoming deal—“here’s the date, here’s the price, here’s the private equity firm”—and at first he just passes it to a friend who is down for the month. Tom explained the fraud triangle: a need (short‑term performance), an opportunity (he could buy up to 0.9% of the fund without approval), and a rationalization (“These other guys are doing it… I’ll do it just this one time”). Across four illegal trades, he personally made just $46,000, which he calls “the price of professional suicide.” The real draw wasn’t the money; it was the illusion of being “on the inside,” part of the in‑group he’d envied since his Wharton days.
Source: Computational Materials Science, Volume 266