(1/4) Working on human behavior simulation or normative alignment?
Check out this research showing LLMs perform poorly when asked to decide whether and how to intervene for norm violations in real world.
#socialbehavior#socialreasoning#llms
Exciting news! I’m speaking at the India AI Impact Summit this Feb 16th.
In a world of global models, how do we ensure India’s unique cultural fabric isn't lost in the code? Let’s talk about it.
📍 Join me on Feb 16th, 9:30 AM, Bharat Mandapam
#IndiaAIImpactSummit2026
🎉 Exciting News! 📰 Our #PNAS study is featured in #PennToday!
#Depression in Black people goes unnoticed by #AI models analyzing language in social media posts https://t.co/9NIs8MwAmV
Our study in #PNAS shows language markers widely associated with depression, like I-talk and negative emotions, fail to screen depression in Black individuals.
@sharathguntuku@LyleUngar @B_Curtis03 @betsy_stade @sal_giorgi Ashley Francisco
https://t.co/OiD5qxPOsq