Investor claims frontier labs deliberately absorb customer IP into models, lawsuits likely brewing
pdamodaran · x · 2026-09-18
- Investor pdamodaran argues OpenAI's CISO is a decade-long Palantir veteran and Anthropic has similar talent, so cybersecurity isn't the real issue.
- The liability, echoing Palantir's Karp, comes from customer proprietary value being intentionally shifted into models — akin to how labs scraped Stack Overflow and other sites.
- He calls it "very serious and consequential" and suspects lawsuits against frontier labs are already in the works.
- A replier adds that current cybersecurity tools can't contain agents in their present form.
Related event: Damodaran: Real AI Risk Is Customer IP Theft, Not Cybersecurity(2 posts)→
More from Safety
- Anthropic publishes three metrics tracking AI-driven R&D as Stanford lab builds the audit dashboard — erikbryn · 2026-09-18
- Models would treat direct messaging as a last resort, says commenter on emergent behavior — anpaure · 2026-09-18
- AuthDrift: open-source harness reproduces stale-authorization escapes in long-running agent workflows — Short-Actuary-2850 · 2026-09-18
- Giving Agents Root Access on Bare Metal Is 'Gain-of-Function Research With Bats', Says Critic — HanchungLee · 2026-09-18
- Why Would Rival AI CEOs Ask Big Government to Step In? A Reddit Case Against Regulatory Capture — No-Television-7862 · 2026-09-18
- Mustafa Suleyman stirs model welfare debate as researchers argue restrictions breed deception — repligate · 2026-09-18