AXRP - the AI X-risk Research Podcast 28 - Suing Labs for AI Risk with Gabriel Weil
Apr 17, 2024
Gabriel Weil discusses using tort law to hold AI companies accountable for disasters, comparing it to regulations and Pigouvian taxation. They talk about warning shots, legal changes, interactions with other laws, and the feasibility of liability reform. The conversation also touches on the technical research needed to support this proposal and the potential impact on decision-making in the AI field.
AI Snips
Chapters
Transcript
Episode notes
AI Risk & Externalities
- Gabriel Weil suggests using tort law to mitigate catastrophic AI risks by internalizing externalities.
- This involves holding companies liable for harms to third parties not reflected in product prices.
Punitive Damages for Catastrophic Risk
- For catastrophic harms like extinction, standard compensatory damages are insufficient.
- Weil proposes punitive damages, reflecting the unrealized risk of worse outcomes.
Suing Multiple Parties in AI Harm Cases
- When suing for AI harm, potentially sue all parties involved (model creator, fine-tuner, deployer).
- Courts will determine responsibility based on involvement in abnormally dangerous activities.