Farnaz Jahanbakhsh

What would it look like to move beyond normatively imposing a single set of values on everyone, and instead build generative AI systems that adapt to the values that matter most to each person in each situation? Excited that our work led by @rashonpoole.bsky.social will be at EMNLP!

📰 Our piece with @ziv-e.bsky.social and @mbernst.bsky.social in The Conversation today. Feed algorithms optimize for engagement behavior because it is easy to measure. The result is that the values amplified in people's feeds are misaligned with what they value. theconversation.com/why-social-m...

Do our social media algorithms correctly reflect our values? Our new article published today in @pnas.org shows that the answer is often not, and that the content that gets promoted into their ranked feeds is often actively counter to our values.

Pleased to share our new paper forthcoming in @icwsm.bsky.social! We introduce a novel framework to measure value expressions in social media posts at scale, leveraging personalization to handle the inherent subjectivity of human values. arxiv.org/abs/2511.08453

Our new CHI paper raises two connected ideas: (1) we can’t—and shouldn’t expect to—agree on all accounts of harm, which makes universal moderation inherently limited; and (2) instead of suppressing the content that harms a given user, we can transform it in a personalized way to preserve its value.

Our new article in @science.org enables social media reranking outside of platforms' walled gardens. We add an LLM-powered reranking of highly polarizing political content into N=1256 participants' feeds. Downranking cools tensions with the opposite party—but upranking inflames them.

📣 Yale workshop, Oct 16-17! 📣 How could/should content ranking work? What's new in content moderation? How can platforms promote civility? Hosted by Yale's Institute for Foundations of Data Science (FDS). Great speakers! Submit posters by 9/22! Spread the word! yalefds.swoogo.com/socialalgori...

📢📢I'm recruiting PhD students for Fall 2025 to join my research lab at the University of Michigan! Interested in doing research in social computing systems? Consider applying! 📢📢 Application deadline: Dec 15 cse.engin.umich.edu/academics/gr...

New paper: Do social media algorithms shape affective polarization? We ran a field experiment on X/Twitter (N=1,256) using LLMs to rerank content in real-time, adjusting exposure to polarizing posts. Result: Algorithmic ranking impacts feelings toward the political outgroup! 🧵⬇️

My work was featured in the MIT News today! Read more on empowering users to assess the accuracy of content and see assessments from their trusted sources everywhere on the web: news.mit.edu/2024/new-too...