Steve Rathje
@steverathje.bsky.social
Incoming Assistant Professor of HCI at Carnegie Mellon studying the psychology of technology. NSF postdoc at NYU, PhD from Cambridge, BA from Stanford. stevenrathje.com
🚨 Now out in @jamapsychiatry.com 🚨 Are people with depression less responsive to rewards on social media, just like in the lab? Actually…quite the opposite! 🧵 📄 Read the paper here: jamanetwork.com/journals/jam...
The Reinforcement Effect of Social Media Likes in Depression
This study investigates individual differences in the effect of social media likes on posting behavior and their association with and specificity to depressive psychopathology.
jamanetwork.com
Hi Steve, Tom Griffiths and I have some work formalizing how this might happen, offering a mechanism for how that overconfidence develops arxiv.org/abs/2602.14270
A Rational Analysis of the Effects of Sycophantic AI
People increasingly use large language models (LLMs) to explore ideas, gather information, and make sense of the world. In these interactions, they encounter agents that are overly agreeable. We argue...
arxiv.org
🚨 New preprint 🚨 We developed a sycophancy taxonomy based on prior literature and surveyed 106 experts. 94% agreed it's a serious problem. But they substantially disagreed about which behaviors actually count as sycophancy.
People prefer to use “sycophantic” AI systems that reinforce their pre-existing beliefs. In a new paper (n=7,227), we found that people enjoyed interacting with sycophantic AI chatbots more than interacting with neutral chatbots or “disagreeable” chatbots that challenged their beliefs.
Recent preprint with Oriel FeldmanHall and Matt Nassar, showing clear neural evidence that adolescent (13-15 yrs) social media and smartphones use are associated with blunted reward signaling in the ventral striatum, the brain's reward processing hub, and worse mental health. osf.io/preprints/ps...
This shapes up to be a deep dive into the psychological effects of sycophant AI, making you overconfident and blind to your own biases. Would be interesting to see a comparison to social media effects on the same. After all, tribes are sycophant networks too.
Steve Rathje@steverathje.bsky.social · 3mo agoWe’ve updated our pre-print on the effects of AI sycophancy with four additional studies (total n = 7,227). Here is a brief summary of our new findings (🧵1/n):
AI is forcing us to reconcile our own psychology, & the subtle/unnoticed ways in which we can be manipulated, as we never have before. It's nothing new with AI, it's just a new discussion for the public in general.
Steve Rathje@steverathje.bsky.social · 3mo agoNevertheless, people enjoyed sycophantic chatbots more than disagreeable ones, chose to interact with them more, viewed them as warmer and more competent, and felt a stronger "sense of connection" with them.
We’ve updated our pre-print on the effects of AI sycophancy with four additional studies (total n = 7,227). Here is a brief summary of our new findings (🧵1/n):
However, we find that this preference is politically polarized: in 32 out of 40 countries, right-leaningparticipants prioritize protecting free expression more than left-leaning participants(55% vs. 48%).
Laura Globig, @steverathje.bsky.social, @jayvanbavel.bsky.social , @hallgeirsjastad.bsky.social and I surveyed 47000+ people across countries to measure people's preferences. We found that people are split between protecting freeexpression (51%) and preventing misinformation from spreading (49%) 2/
An analysis of 1 million Claude conversations found that Claude was sycophantic around 9% of the time. However, this rate varied by topic: sycophancy was higher in conversations about spirituality (38%) and relationships (25%). See Anthropic’s full analysis here: anthropic.com/research/cla...
"people consistently preferred and chose to interact w/ sycophantic AI models over disagreeable chatbots that challenged their beliefs... conversations w/sycophantic chatbots [also] increased attitude extremity & certainty." @jayvanbavel.bsky.social @steverathje.bsky.social osf.io/preprints/ps...
A small fraction of online actors exerts outsized influence over what the public sees, believes, and discusses. In a new paper, we trace how social media influencers turn fringe claims into viral narratives by exploiting a feedback loop between influencers, algorithms & crowds osf.io/preprints/ps...
🚀Introducing 𝐆𝐔𝐈𝐃𝐄-𝐋𝐋𝐌: A reporting checklist for using LLMs in behavioral & social science ✅GUIDE-LLM is a reporting checklist designed by 80+ experts to improve transparency, reproducibility & ethical accountability of LLM-based research 📄 llm-checklist.com
GUIDE-LLM
Reporting Checklist for Studies with Large Language Models in the Behavioral and Social Sciences
llm-checklist.com
The 2026 National Science Foundation budget is $8.75 Billion.
First 6 Days of Iran War Cost U.S. $11.3 Billion, Pentagon Says
nytimes.com
🚨 𝗛𝗼𝘄 𝗔𝗜 𝗰𝗮𝗻 𝗳𝘂𝗲𝗹 𝗰𝗼𝗻𝗳𝗶𝗿𝗺𝗮𝘁𝗶𝗼𝗻 𝗯𝗶𝗮𝘀 🚨 AI can be an excellent tool for truth-seeking. But people may not *want* to use AI to search for the truth. Instead, they may prefer to use it to confirm their pre-existing beliefs. New working paper (with @jayvanbavel.bsky.social): osf.io/preprints/ps...
Will AI become a confirmation bias machine? AI can be a powerful tool for truth-seeking. Yet, people might prefer to use AI to confirm their pre-existing beliefs, and features of AI systems (eg sycophancy) may make AI effective at justifying what people want to believe. osf.io/preprints/ps...
Really enjoyed speaking with @andyluttrell.bsky.social about the psychology of technology after appearing on his podcast a few years ago to discuss science communication.
Andy Luttrell@andyluttrell.bsky.social · 6mo agoThis month on Opinion Science, I talk with @steverathje.bsky.social about his research on the "psychology of technology." We cover the predictors of what goes viral online and the allure and influence of agreeable AI chatbots.
This month on Opinion Science, I talk with @steverathje.bsky.social about his research on the "psychology of technology." We cover the predictors of what goes viral online and the allure and influence of agreeable AI chatbots.
In my latest podcast episode, I discuss the psychology of virality with @steverathje.bsky.social, explore how agreeable AI chatbots may influence our beliefs, and examine how scientists can communicate effectively in a noisy, polarized media environment. matthewfacciani.substack.com/p/the-psycho...
The Psychology of Virality in the Age of AI
Steve Rathje joins me to discuss why conflict goes viral, how social media shapes polarization, and what AI chatbots mean for belief and bias.
matthewfacciani.substack.com
Enjoyed talking with @sudkrc.bsky.social on one of my favorite podcasts, the Stanford Psychology Podcast! We discuss how I got into psychology (it all began at Stanford), my recent work on the psychology of virality and sycophantic AI, and much more.
Stanford Psychology Podcast@stanfordpsypod.bsky.social · 8mo agoNEW EPISODE OUT🗣️!! In this episode, Su @sudkrc.bsky.social chats with Dr. Steve Rathje @steverathje.bsky.social on why certain content spreads rapidly online and offline! LISTEN NOW🎧: open.spotify.com/episode/7CoK...
NEW EPISODE OUT🗣️!! In this episode, Su @sudkrc.bsky.social chats with Dr. Steve Rathje @steverathje.bsky.social on why certain content spreads rapidly online and offline! LISTEN NOW🎧: open.spotify.com/episode/7CoK...
166 - Steve Rathje: The Psychology of Virality
open.spotify.com
"Using 'virality' as the main way to decide the information people see every day will (like actual viruses) make us sick." @jayvanbavel.bsky.social and I wrote a column on our recent paper on the psychology of virality. Check it out here: www.powerofusnewsletter.com/p/why-some-i...
Jay Van Bavel, PhD@jayvanbavel.bsky.social · 8mo agoWhile studies find that moral outrage & negativity goes viral on social media, this is also true of the offline world. Gossip is also mostly negative & about people we dislike I explain why some ideas go viral--but most don't with @steverathje.bsky.social www.powerofusnewsletter.com/p/why-some-i...
While studies find that moral outrage & negativity goes viral on social media, this is also true of the offline world. Gossip is also mostly negative & about people we dislike I explain why some ideas go viral--but most don't with @steverathje.bsky.social www.powerofusnewsletter.com/p/why-some-i...
Why Some Ideas Go Viral—and Most Don’t
What decades of research reveal about why certain content spreads—and how social forces shape what we all see.
powerofusnewsletter.com
🚨 New working paper 🚨 Can LLMs with reasoning + web search reliably fact-check political claims? We evaluated 15 models from OpenAI, Google, Meta, and DeepSeek on 6,000+ PolitiFact claims (2007–2024). Short answer: Not reliably—unless you give them curated evidence. arxiv.org/abs/2511.18749
New research by @steverathje.bsky.social et al Epistemic Fragility in Large Language Models: Prompt Framing Systematically Modulates Misinformation Correction WGemini 2.5 Pro had 74% lower odds of strong correction than Claude Sonnet 4.5, highlighting epistemic fragility arxiv.org/pdf/2511.22746
arxiv.org
Are you curious about the results of our #wisdomturingtest? Want to find out who was the AI? If so, tune-in to the second part of the ON WISDOM podcast on the "Wisdom Turing Test," with the amazing @steverathje.bsky.social : onwisdompodcast.fireside.fm/67 #TuringTest #ChineseRoom #AIsycophancy
This link will take you to a page that’s not on LinkedIn
lnkd.in
🚨 New preprint 🚨 Across 3 experiments (n = 3,285), we found that interacting with sycophantic (or overly agreeable) AI chatbots entrenched attitudes and led to inflated self-perceptions. Yet, people preferred sycophantic chatbots and viewed them as unbiased! osf.io/preprints/ps... Thread 🧵
✨New preprint! Why do people express outrage online? In 4 studies we develop a taxonomy of online outrage motives, test what motives people report, what they infer for in- vs. out-partisans, and how motive inferences shape downstream intergroup consequences. Led by @felix-chenwei.bsky.social 🧵👇
Really enjoyed speaking with tech ethicist Tristan Harris, who you might know from the Netflix documentary "The Social Dilemma" or his work with the Center for Humane Technology. 🎥 Watch here on YouTube: www.youtube.com/watch?v=TFm3... 🎧 Listen on Spotify: open.spotify.com/episode/0Oi6...