←
James Landay21,253 posts
James Landay avatar

James Landay

@landay

Professor of Computer Science, Stanford - HCI & Design. Co-founder & Director @StanfordHAI. Personal opinions, not Stanford's, jlanday.bsky.social

◷ Joined March 2007▧ 441 Photos and videos
James Landay avatar
James Landay@landay·Nov 3

Hope folks at #UIST26 got to see this cool talk by @sapkotashardul_ on why AI should understand our long-term goals and how we might shape that understanding. programs.sigchi.org/uist/202…

AI systems understand what we’re doing right now. What if they could also understand what we’re working toward long term and why it matters to us? We built Tempo to connect computer use to long-term goals, with inferences you can inspect and correct. Our UIST 2026 paper 🧵

0:30
Stanford HAI avatar
⟳ James Landay retweeted
Stanford HAI@StanfordHAI·Oct 5

Benchmark scores influence which AI models get funded, bought, and regulated. Two new studies from Stanford researchers and collaborators, supported by Stanford HAI, ask whether those tests measure what they claim to. Read more: hai.stanford.edu/news/the-te…

James Landay avatar
James Landay@landay·Sep 29

It is important for AI to understand our long-term goals and help us *achieve* these goals. Important work from a great PhD student I work with @sapkotashardul_. Supported by @StanfordHAI and @HPI_DE. Check it out!

AI systems understand what we’re doing right now. What if they could also understand what we’re working toward long term and why it matters to us? We built Tempo to connect computer use to long-term goals, with inferences you can inspect and correct. Our UIST 2026 paper 🧵

0:30
James Landay avatar
James Landay@landay·Sep 23

Parker Ruth, a PhD student I advise @Stanford, devised an improved method to analyze heart signals. His Bayesian pulse deconvolution method helps extract accurate information from health device biosignals. Read about it in our npj Digital Medicine paper. hci.st/bayesianpd

Post attachment
Stanford HAI avatar
⟳ James Landay retweeted
Stanford HAI@StanfordHAI·Sep 16

Downloading a model is increasingly easy. Understanding how it was made — or changing it at its root — is another matter. “Open weights answer ‘Can I run this?’ Open source answers ‘Can I trust this, improve it, and build the next thing on top of it?’ Right now almost everyone is answering the first question but nowhere close to the second.” Stanford HAI Denning Director @landay, cited in The Register: theregister.com/columnists/2…

James Landay avatar
James Landay@landay·Sep 14

From ImageNet for computer vision to HELM for LLMs, Stanford’s AI team has been at the forefront of open source evaluation of AI. @StanfordHAI, home to @StanfordAILab, supports researchers at the forefront of AI model benchmarking. It takes a village to shepherd great technology!

I propose Stanford NLP as an independent third-party evaluator under @DarioAmodei’s 3 step plan. For important parts of the work, universities would be better than any other organization (see below 🧵👇), and, of university groups, @stanfordnlp would be the best one to choose. 😊

James Landay avatar
James Landay@landay·Apr 19

Important work!

Most of what I actually need help with, I never think to tell a model. But why is it on me to remember? Our new paper asks: what if AI could proactively specialize to individuals and the tasks they’re carrying out at this very moment? 🧵

0:15
Michael Bernstein avatar
⟳ James Landay retweeted
Michael Bernstein@msbernst·Apr 16

Best paper honorable mention at CHI — instantly adaptive objectives for LMs as you use your computer

Most of what I actually need help with, I never think to tell a model. But why is it on me to remember? Our new paper asks: what if AI could proactively specialize to individuals and the tasks they’re carrying out at this very moment? 🧵

0:15
Michelle Lam avatar
⟳ James Landay retweeted
Michelle Lam@michelle123lam·Apr 15

Thanks so much to co-authors @oshaikh13, Hallie Xu, @azguo2, @Diyi_Yang, @jeffrey_heer, @landay, & @msbernst! We’re grateful that this work received a CHI best paper honorable mention! paper: arxiv.org/abs/2510.14591 waitlist: forms.gle/rVzWo5asJ2Xv6QEq7 talk: Th 4/16 12pm @ P1-111

James Landay avatar
James Landay@landay·Apr 16

Star co-advised student @michelle123lam giving a talk on very cool work this morning at #chi2026 in room 111 at noon. Come see it! Best paper honorable mention winner.

Most of what I actually need help with, I never think to tell a model. But why is it on me to remember? Our new paper asks: what if AI could proactively specialize to individuals and the tasks they’re carrying out at this very moment? 🧵

0:15
James Landay avatar
James Landay@landay·Apr 15

Next session: 🚨 Best Paper at #CHI2026 🚨 Don’t miss “Bloom: Designing for LLM-Augmented Behavior Change Interactions” by my co-advised student Matthew Jörke (@mjoerke), exciting work on how LLMs can support real behavior change. stanfordhci.github.io/Bloom/ 11:39am (Wed), Room 111