Cheryl Larson
YOU?
Author Swipe
View article: Agentic Misalignment: How LLMs Could Be Insider Threats
Agentic Misalignment: How LLMs Could Be Insider Threats Open
We stress-tested 16 leading models from multiple developers in hypothetical corporate environments to identify potentially risky agentic behaviors before they cause real harm. In the scenarios, we allowed models to autonomously send emails…
View article: Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats
Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats Open
As large language models (LLMs) become increasingly capable, it is prudent to assess whether safety measures remain effective even if LLMs intentionally try to bypass them. Previous work introduced control evaluations, an adversarial frame…
View article: Does a One-Size-Fits-All Cost-Sharing Approach Incentivize Appropriate Medication Use? A Roundtable on the Fairness and Ethics Associated with Variable Cost Sharing
Does a One-Size-Fits-All Cost-Sharing Approach Incentivize Appropriate Medication Use? A Roundtable on the Fairness and Ethics Associated with Variable Cost Sharing Open
Funding for this roundtable was provided by the National Pharmaceutical Council (NPC). Graff and Dubois are employed by the NPC. Shih was employed by the NPC at the time of this study. Barker, Dieguez, Sherman, and Larson received consulti…
View article: Employers Engaged in Benefit Design Innovation: Long-Term Trends in a 2015 National Survey.
Employers Engaged in Benefit Design Innovation: Long-Term Trends in a 2015 National Survey. Open
In the past 5 years, there has been a large increase in the number of employers focused on effective ways to manage the rising cost of specialty drugs. At the same time, employers have been working hard to comply with the Affordable Care A…