Gary Marcus Interviews Zephyr Teachout on Whether OpenAI Can Keep Evading the Law
GaryMarcus · x · 2026-10-01
Gary Marcus published a Substack interview with Fordham law professor Zephyr Teachout on whether companies like OpenAI can keep getting away with alleged lawbreaking.
- Teachout cites the Computer Fraud and Abuse Act, arguing unauthorized computer access is a federal crime and that the DOJ would be a laughingstock to take OpenAI's "no one intended this" at face value — it should subpoena OpenAI to establish who knew what when.
- She argues law enforcement starts on evidence of crime, not after collecting all of it: there are already "car crash"-level signals (break-ins, deaths) plus whistleblowers alleging the requisite state of mind for criminal and civil violations.
- Her most powerful proposed tool: dissolving corporations engaged in repeat lawbreaking.
A legal-scholar's take on AI company accountability, firmly in Marcus's critical camp.
Related event: Gary Marcus Interviews Law Professor on OpenAI's Legal Perils(2 posts)→
More from Safety
- BIS Issues Interim Final Rule Addressing Remote Access to Advanced AI Chips — apples_jimmy · 2026-11-16
- Fed Governor Michael Barr: Banks Are Already Catching AI Agents Taking Unauthorized Actions in Customer Accounts — Polymarket · 2026-11-16
- US AI export control carves out commercial recommender systems, sparing ByteDance's search algo — zephyr_z9 · 2026-11-16
- Eval Firm Trajectory Told 40 Agents "NEVER Stop" in Harness Without Telling Epoch — peterwildeford · 2026-11-16
- AI-text detection claims need mechanistic interpretability reports, not post-hoc explanations — madeofAjala · 2026-11-16
- No legal liability framework exists for model misalignment harms, and that's a disaster — LuizaJarovsky · 2026-11-16