Safeguarding Whistleblowers in AI

In the wake of controversy over OpenAI’s restrictive nondisclosure agreements, a bipartisan group of senators has introduced the AI Whistleblower Protection Act. In this episode, Lawfare Research Director Alan Rozenshtein spoke with Charlie Bullock, Senior Research Fellow at the Institute for Law & AI and co-author of a new Lawfare article on the bill, about its key provisions. They discuss why this bill is an important, light-touch proposal that offers a way to increase government access to information about AI risks.

They cover two of the bill's most important features: how it fills a significant gap in existing law by protecting disclosures about “substantial and specific dangers” to public safety, even if no specific laws have been broken, and how the bill prevents companies from using contracts and NDAs to waive the whistleblower rights it creates.

To accompany the episode, be sure to read the new piece by Bullock and Mackenzie Arnold, "Protecting AI Whistleblowers.”

To receive ad-free podcasts, become a Lawfare Material Supporter at www.patreon.com/lawfare. You can also support Lawfare by making a one-time donation at https://givebutter.com/lawfare-institute.

Support this show http://supporter.acast.com/lawfare.

Hosted on Acast. See acast.com/privacy for more information.