MentalHealthBench is an expert-informed benchmark for evaluating helpful and safe AI responses across realistic mental health conversations.

Recent efforts in AI ethics have been dominated by the proliferation of high-level principles intended to guide responsible development and deployment. Yet persistent implementation failures suggest…

Abstract. People help those with a reputation for helping others; as a result, they are more likely to behave generously when reputational concerns are pre

Rather than racing to replace people, we should build AI tools that amplify human expertise and expand opportunity.

Abstract. Chatbots are now able to engage in sophisticated conversations with consumers in the domain of relationships, providing a potential coping soluti

"Virtue is not completed in reflection; it is completed in life. The model never comes down the mountain; its entire existence is the conversation. There is no worl…

New Report: Towards Open Benchmarks for Human Flourishing with AIMost AI benchmarks measure how smart models are—but not how they affect us. Do they deep…

A three-year, US$3.6M project co-led by SRI Research Leads Karina Vold and Ashton Anderson with Purdue University’s Louis Tay explores how AI systems might be designed to support long-term human well-...

Recent advances in AI have enabled large language models to produce expressions that seem empathetic to human users, raising scientific and ethical questions about how people perceive and choose…