Columbia law professor Neel Guha to discuss evaluating legal AI at Crosby Intelligence event
graceisford · x · 2026-11-11
Crosby Intelligence hosts Columbia Law professor Neel Guha (creator of the 162-task LegalBench benchmark) in conversation with research engineer Sharan Ramjee on Nov 17.
Topics: how to evaluate models in a domain where expert lawyers disagree with each other, what benchmarks can and can't tell you about agent readiness for real client work, liability when AI errs, and whether proposed AI regulations (disclosure, registration, licensing, auditing) are technically feasible. His adverse-event-reporting proposals for frontier models have shaped AI policy.
More from Companies & People
- Lenny's late-2026 report: PMs winning, engineers plateauing, designers at record low — lennysan · 2026-11-11
- AI Omakase podcast teases dinner-table conversations with researchers and founders — arankomatsuzaki · 2026-11-11
- Gary Marcus slams Anthropic IPO: 2 shares control 50.1% of votes, retail gets no say — GaryMarcus · 2026-11-11
- UCLA CS opens multiple tenure-track assistant professor searches at $120k-$150k — guyvdb · 2026-11-11
- Anne Griffin to talk open-weight AI models at MLOps World, Austin Nov 17-18 — annetgriffin · 2026-11-11
- Survey of 200 security leaders: security sets the pace of enterprise AI adoption — BenBajarin · 2026-11-11