POLITICO Morning Tech today highlighted a new policy brief co-authored by Federal Alumni Fellows Stephanie Nugyen and Erie Meyer. The piece, titled What Would It Take for AI Companies to Reduce AI Sycophancy Risks?, outlines concrete steps companies can take to curb the risks of overly flattering or agreeable AI systems – particularly when users are struggling with mental health issues.

Nguyen and Meyer co-authored the brief with Samuel A. Levine, Senior Fellow at UC Berkeley Center for Consumer Law & Economic Justice. The authors urge AI companies to adopt rigorous safety standards, separate revenue decisions from safety oversight, tie safety outcomes to employee performance, and enable independent audits of generative AI models.

The full brief, published by Georgetown’s Institute for Technology Law & Policy, is available here:

What Would It Take for AI Companies to Reduce AI Sycophancy Risks? | Institute for Technology Law & Policy | Georgetown Law(This link opens in a new tab)