Sign in

Kyra Wilson

@kyrawilson.bsky.social
98 followers 164 following 24 posts

PhD student at UW iSchool | ai fairness, evaluation, and decision-making | she/her 🥝 kyrawilson.github.io/me

PostsRepliesMedia
Kyra Wilson @kyrawilson.bsky.social · 23/04/2026
Thank you Adam!!!
010
Kyra Wilson @kyrawilson.bsky.social · 23/04/2026
I am so honored to have been chosen as a Scholar! I couldn’t have done it without the help of my community at UW and beyond. Incredibly grateful to everyone, and equally excited to keep going and do some more great research! 🏆🙏✨
070
Kyra Wilson @kyrawilson.bsky.social · 21/11/2025
We know AI is biased, but what does that do to people's hiring decisions? (Spoiler alert: nothing good) To address this, @aylincaliskan.bsky.social and I wrote about legal implications and policy suggestions to protect people's rights and autonomy @brookings.edu www.brookings.edu/articles/ais...
urldefense.com
AI's threat to individual autonomy in hiring decisions | Brookings
Kyra Wilson and Aylin Caliskan discuss new research on AI's influence in decision-making and what this portends for policymakers.
081
Reposted by Kyra Wilson
Michael Saxon @saxon.me · 19/11/2025
And here is the presentation I gave on networking, self-promo, and how to make the most out of a conference. Hope this helps for everyone at NeurIPS! www.youtube.com/watch?v=B9hG...
youtube.com
Conferencemaxxing: How to grow your profile and network as a scientist
YouTube video by Michael Saxon (NLP & Generative AI research)
0146
Reposted by Kyra Wilson
Your City @yourcityseattle.bsky.social · 13/11/2025
We took on a powerful incumbent who was expected to coast to reelection. We faced more corporate PAC money than has ever been spent attacking a candidate in a Seattle election. We built a people-powered movement rooted in hope for our city’s future. And we won. This is YOUR city!
A picture of Katie Wilson on election night at her campaign party delivering a speech to supporters. she holds her hand over her heartA picture of Katie Wilson at her election party on election night. speaking into a microphone to a crowd of supporters, there are blue balloons behind her on stage and gold and silver streamersA black and white photo of Katie Wilson knocking on a door of a single family home while door knocking for her campaignA photo of Katie Wilson looking at her phone while standing by the buzzer box for an apartment building while door knocking for her campaign
441365216
Kyra Wilson @kyrawilson.bsky.social · 14/11/2025
Thanks for shout-out Samer, I’m glad our work resonated with you!
010
Reposted by Kyra Wilson
Samer Hassan @samerp2p.bsky.social · 13/11/2025
"No Thoughts Just AI: Biased LLM Hiring Recommendations Alter Human Decision Making and Limit Human Autonomy" by @kyrawilson.bsky.social et al It shows how people trust AI hiring proposals blindly, even when the AI's being racist. This highlights the importance to tackle algorithmic bias
211
Kyra Wilson @kyrawilson.bsky.social · 10/11/2025
Thanks to @uwnews.uw.edu for covering my + @aylincaliskan.bsky.social's recent work published at AIES 2025! www.washington.edu/news/2025/11...
washington.edu
People mirror AI systems’ hiring biases, study finds
In a new UW study, 528 participants worked with simulated AI systems to select job candidates. The researchers simulated different levels of racial biases for resumes from white, Black, Hispanic and.....
0107
Reposted by Kyra Wilson
Kshitish Ghate @kghate.bsky.social · 14/10/2025
🚨New paper: Reward Models (RMs) are used to align LLMs, but can they be steered toward user-specific value/style preferences? With EVALUESTEER, we find even the best RMs we tested exhibit their own value/style biases, and are unable to align with a user >25% of the time. 🧵
1127
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
If you’ve made it to the end of this thread, thanks for reading! I hope we can connect soon and chat about the role of all these works in making AI evaluation more valid, reliable, and actionable for the real-world! 🥳
000
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
I will present this tomorrow at 11:45am in Paper Session 6: Integrating AI into the Workplace. The preprint has more analyses of AI literacy's impact and discussion of the implications of this work for human autonomy in high-risk domains. www.arxiv.org/abs/2509.04404
arxiv.org
No Thoughts Just AI: Biased LLM Hiring Recommendations Alter Human Decision Making and Limit Human Autonomy
In this study, we conduct a resume-screening experiment (N=528) where people collaborate with simulated AI models exhibiting race-based preferences (bias) to evaluate candidates for 16 high and low st...
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
But a positive note is that when people took an implicit association test (commonly used for anti-bias training) before doing the resume-screening task, they increased their selection of stereotype-incongruent candidates by 12.7% regardless of how biased the AI model they interacted with was.
An image showing that people who completed an implicit association test before the resume-screening task were more likely to select Black and Hispanic candidates for high-status jobs and Asian candidates for low status jobs.
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
We showed people AI recommendations that had varying levels of racial bias and found that (at most) human oversight decreased bias in final outcomes by at most 15.2% which is still far from the outcome bias rates when no AI or unbiased AI was used.
An image showing that people's decisions are very closely aligned with biased AI recommendations. Only in the most severe cases of AI bias do people make less biased decisions.
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
3️⃣ Last but not least, I’ll be talking about a human-subjects experiment follow-up to my AIES 2024 paper with @aylincaliskan.bsky.social, also co-authored by Mattea Sim and Anna-Maria Gueorguieva! arxiv.org/abs/2407.20371
arxiv.org
Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval
Artificial intelligence (AI) hiring tools have revolutionized resume screening, and large language models (LLMs) have the potential to do the same. However, given the biases which are embedded within ...
120
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
I'll be discussion this work tomorrow at 3:15pm during Poster Session 4! Our preprint has additional findings and analyses comparing generated images to human baselines and measuring dimensions of skin tone other than lightness/darkness. arxiv.org/abs/2508.17465
arxiv.org
Bias Amplification in Stable Diffusion's Representation of Stigma Through Skin Tones and Their Homogeneity
Text-to-image generators (T2Is) are liable to produce images that perpetuate social stereotypes, especially in regards to race or skin tone. We use a comprehensive set of 93 stigmatized identities to ...
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
We also found that depictions of racial identities are getting more homogenized with successive releases of SD, reinforcing harmful ideas about what people with stigmatized identities "should" look like.
An image showing that Stable Diffusion v2.1 generates skin tones which are more diverse for a stigmatized racial identity versus Stable Diffusion XL which generates darker, more homogenous skin tones.
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
We found that the newest model (SD XL) tends to generate images with darker skin tones compared to SD v1.5 and v2.1, but it still over-represents dark skin tones for stigmatized identities compared to non-stigmatized identities.
An image showing that Stable Diffusion XL has the darkest skin tones on average and stigmatized identities have darker skin tones that non-stigmatized identities.
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
2️⃣ In another work with Sourojit Ghosh and @aylincaliskan.bsky.social, we analyzed skin tones in images of 93 stigmatized identities using three versions of Stable Diffusion (SD).
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
Chat with me about this work at Poster Session 2 (today at 6:15pm) or read our preprint in full! arxiv.org/abs/2508.11067
arxiv.org
Bias is a Math Problem, AI Bias is a Technical Problem: 10-year Literature Review of AI/LLM Bias Research Reveals Narrow [Gender-Centric] Conceptions of 'Bias', and Academia-Industry Gap
The rapid development of AI tools and implementation of LLMs within downstream tasks has been paralleled by a surge in research exploring how the outputs of such AI/LLM systems embed biases, a researc...
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
We also find that 89.4% of papers don’t provide detailed information about real-world implementation of their findings. Based on this, we made a Fact Sheet that to guide researchers in communicating findings in ways that enable model developers or downstream users to implement them appropriately.
An image showing the Fact Sheet we developed, which includes questions for the topics High-level Facts, Limitations and Ethical Considerations, Intended Use Cases, Implementation-level Details, and Maintenance and Lifecycle.
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
1️⃣ Sourojit Ghosh and I conducted a review of AI bias literature and found that 82% do not provide an explicit definition of bias and 79.9% do not explore bias outside of binary gender bias. This means that only a particular subset of marginalized groups may benefit from AI bias research.
100
Kyra Wilson @kyrawilson.bsky.social · 21/10/2025
Happy to share that I’m presenting 3 research projects at AIES 2025 🎉 1️⃣Gender bias over-representation in AI bias research 👫 2️⃣Stable Diffusion's skin tone bias 🧑🏻🧑🏽🧑🏿 3️⃣Limitations of human oversight in AI hiring 👤🤖 Let's chat if you’re at AIES or read below/reach out for details! #AIES25 #AcademicSky
192
Reposted by Kyra Wilson
Shahan Ali Memon @shahanmemon.bsky.social · 27/09/2025
Applying for a #PhD @ischool.uw.edu? Read 👇 Our student-run application feedback program will be open from October 20th through 1st November 2025. Everyone applying, especially those from historically underrepresented groups or who have faced barriers in higher ed are highly encouraged to apply.
157
Reposted by Kyra Wilson
e.w. niedermeyer @niedermeyer.online · 01/08/2025
Tesla is being hit with $329m in damages for a crash in which the human driver says he knew Autopilot wasn't self-driving. This is super important, because it shows that Autopilot's design and marketing can induce inattention even when drivers consciously know they are supposed to pay attention.
261900374
Reposted by Kyra Wilson
Natalia Luka @nataliyan.bsky.social · 21/07/2025
📣 Exceptional resource alert from @ucblaborcenter.bsky.social: a database of U.S. union bargaining provisions related to workplace tech. laborcenter.berkeley.edu/negotiating-... Workers need a voice in how tech is utilized in the workplace— and here are 175 blueprints for what that looks like
ucblaborcenter.bsky.social
UC Berkeley Labor Center (@ucblaborcenter.bsky.social)
The Labor Center conducts research and education on issues related to labor and employment. Our trainings serve to educate a diverse new generation of labor leaders. We also engage UC Berkeley student...
077
Reposted by Kyra Wilson
Prof. Nicole Holliday @mixedlinguist.bsky.social · 13/05/2025
There’s a lot of evil stuff. But these few sentences are a result of wild corruption, and would remove any guardrails against AI systems being used for massive civil and humans rights abuses on an unimaginable scale. Call your congresspeople, and scream about it in the streets.
02910
Reposted by Kyra Wilson
Nelson Niu @nelsonniu.bsky.social · 08/05/2025
I’m in this article along with other @uaw4121.bsky.social members to explain what’s happening to grad student researchers in the US right now. It’s not looking great for us, and it’s looking even worse for America.
032
Reposted by Kyra Wilson
Dr. David Miller 🏳️‍🌈 @davidimiller.bsky.social · 18/04/2025
So: In 2017, Congress made this happen: NIH: Proposed 22% cut --> 9% increase NSF: Proposed 11% cut --> 4% increase NOAA: Proposed 16% cut --> 4% increase Obviously, 2025 is not 2017. A lot is diff now. But still: 💰 Congress, not WH, sets budgets. 📞 Public support & calls to Congress matter.
219699
Reposted by Kyra Wilson
Gautam Kamath @gautamkamath.com · 01/05/2025
I wrote a post on how to connect with people (i.e., make friends) at CS conferences. These events can be intimidating so here's some suggestions on how to navigate them I'm late for #ICLR2025 #NAACL2025, but in time for #AISTATS2025 #ICML2025! 1/3 kamathematics.wordpress.com/2025/05/01/t...
kamathematics.wordpress.com
Tips on How to Connect at Academic Conferences
I was a kinda awkward teenager. If you are a CS researcher reading this post, then chances are, you were too. How to navigate social situations and make friends is not always intuitive, and has to …
36618
Reposted by Kyra Wilson
Kshitish Ghate @kghate.bsky.social · 29/04/2025
Excited to announce our #NAACL2025 Oral paper! 🎉✨ We carried out the largest systematic study so far to map the links between upstream choices, intrinsic bias, and downstream zero-shot performance across 131 CLIP Vision-language encoders, 26 datasets, and 55 architectures!
1216
Reposted by Kyra Wilson
Sarah Quinn @sarahquinn.bsky.social · 28/04/2025
Faculty at the University of Washington, please sign, and please help us spread the word! Link the letter is below.
dailyuw.com
Hundreds of UW faculty members sign open letter opposing staff layoffs in College of Arts & Sciences
Approximately 270 UW faculty members have signed an open letter condemning a proposed plan to lay off staff in the College of Arts & Sciences (CAS), calling the decision “deeply
123
Kyra Wilson @kyrawilson.bsky.social · 25/04/2025
🤩 If you made it this far, thanks for reading! Be sure to also check out our research paper which quantified the risks these systems pose to different gender, race, and intersectional groups! (6/6) arxiv.org/abs/2407.203...
arxiv.org
Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval
Artificial intelligence (AI) hiring tools have revolutionized resume screening, and large language models (LLMs) have the potential to do the same. However, given the biases which are embedded within ...
010
Kyra Wilson @kyrawilson.bsky.social · 25/04/2025
✅ Applicants should also be aware of when AI is being used to screen their materials and if it was used for adverse actions. These rules already apply for things like background checks, why not AI screening too? Let's empower people by making AI screening consensual and transparent! (5/6)
120
Kyra Wilson @kyrawilson.bsky.social · 25/04/2025
🔀 We also found that audits looking at gender and race in isolation might not capture the true societal harms for intersectional groups. We need to clarify hiring discrimination regulations to account for intersectional harms and apply those standards to audits to understand risk and impact. (4/6)
110
Kyra Wilson @kyrawilson.bsky.social · 25/04/2025
🏛 There has been some great legislation coming out of New York City and Colorado requiring auditing of these tools. These could be improved by empowering third parties to conduct audits instead of relying on self-reporting and not exempting systems with humans in-the-loop! (3/6)
110
Kyra Wilson @kyrawilson.bsky.social · 25/04/2025
⚠️ Current approaches to debiasing models aren't enough! We still don't know how protected characteristics are signaled in resumes or used by automatic screening tools, so we will need help from regulators, model developers, and users to make these tools work for everyone. (2/6)
110
Kyra Wilson @kyrawilson.bsky.social · 25/04/2025
🗞️ Hot off the press! 🗞️ @aylincaliskan.bsky.social and I wrote a blog post about how to make resume screening with AI more equitable based findings from our work presented at AIES in 2024. Major takeaways ⬇️ (1/6) www.brookings.edu/articles/gen...
brookings.edu
Gender, race, and intersectional bias in AI resume screening via language model retrieval
Kyra Wilson and Aylin Caliskan examine gender, race, and intersectional bias in AI resume screening and suggest protective policies.
164
Reposted by Kyra Wilson
Liz Neeley @lizneeley.bsky.social · 08/04/2025
Urgently reposting! University leadership need to actively check the SEVIS database every day (studyinthestates.dhs.gov/sevis-help-hub). Ask your chair to commit to it. Yesterday, UW discovered 9 visa cancellations & only found out b/c they ran a status check. www.seattletimes.com/seattle-news...
seattletimes.com
UW students’ visas revoked without notice by Trump administration
The University of Washington plans to provide the affected students access to legal services that are paid for by their student fees.
1132108
Reposted by Kyra Wilson
Tech Policy Lab @techpolicylab.bsky.social · 01/04/2025
Co-Director @aylincaliskan.bsky.social and Faculty Associate @emilymbender.bsky.social were recognized by @brookings.edu for being an important (un)Hidden Figures in AI. www.brookings.edu/articles/rec...
brookings.edu
Recognizing the (un)Hidden Figures in AI
The AI Equity Lab highlights leading voices and figures in artificial intelligence in recognition of Women's History Month.
021
Reposted by Kyra Wilson
Shannon Mattern @shannonmattern.bsky.social · 30/03/2025
“Many of these men fancied themselves to be a cut above the [women] programmers who came before, and they often perceived COBOL as inferior and unattractive, in part because it did not require abstruse knowledge of underlying computer hardware or a computer science qualification.”
logicmag.io
Built to Last
Don’t blame COBOL. Blame austerity.
13555160
Reposted by Kyra Wilson
Michael Yarbrough (he/him) @mwyarbrough.bsky.social · 30/03/2025
This exciting resolution from Rutgers University Senate is exactly the kind of model I think higher ed needs: a mutual defense pact of, in this case, Big 10 schools. They call for member institutions to create a joint defense fund and make resources available to any member institution under attack.
Excerpt from Rutgers University Senate resolution calling for a Mutual Academic Defense Compact. The excerpt reads:

"BE IT RESOLVED THAT, the Rutgers University Senate urges the President of Rutgers University to formally propose and help establish a Mutual Academic Defense Compact (MADC) among all members of the Big Ten Academic Alliance;

BE IT FURTHER RESOLVED THAT, under this compact, all participating institutions shall commit meaningful funding to a shared or distributed defense fund. This fund shall be used to provide immediate and strategic support to any member institution under direct political or legal infringement;"
6646971527
Reposted by Kyra Wilson
Aylin Kamelia Caliskan @aylincaliskan.bsky.social · 19/02/2025
UW’s @techpolicylab.bsky.social and I invite applications for a 2-year Postdoctoral Researcher position in "AI Alignment with Ethical Principles" focusing on language technologies, societal impact, and tech policy. Kindly share! apply.interfolio.com/162834 Priority review deadline: 3/28/2025
apply.interfolio.com
Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio
01010
Reposted by Kyra Wilson
UAW Local 4121 @uaw4121.bsky.social · 18/02/2025
Show up to tell the Trump administration: our research and our jobs need funding, not freezes! RSVP here: www.labor4highered.org
labor4highered.org
044
Reposted by Kyra Wilson
Eric Blanc @ericblanc.bsky.social · 12/02/2025
One concrete thing everybody can do to support federal workers is participate in (and spread the word about) the February 19 Save Our Services day of action bit.ly/SOSaction
024394
Reposted by Kyra Wilson
Adam Visokay @avisokay.bsky.social · 26/01/2025
The Responsibility in AI Systems & Experiences at UW is excited to host a Winter research exhibition on Feb 29 in Seattle. Info about our keynote speaker, accepted posters and a link to RSVP are on the event page now! www.raise.uw.edu/winterexposi...
011
Reposted by Kyra Wilson
Kush Varshney कुश वार्ष्णेय @krvarshney.bsky.social · 11/12/2024
Nice article explaining unwanted bias in hiring stemming from the use of artificial intelligence with extensive quotations from @kyrawilson.bsky.social and Moninder Singh. www.ibm.com/think/news/a...
ibm.com
How AI can help the recruitment process—without hurting it | IBM
AI agents for recruiters are time-savers—but experts caution that developers need to make sure they're unbiased. Find out what data scientists and engineers are doing to mitigate bias in AI-driven rec...
083
Reposted by Kyra Wilson
ThisDay @thisday.bsky.social · 03/11/2024
#geekwire - #AI overwhelmingly prefers white and male job candidates in new test of resume-screening bias #Blackmen fared the worst with the models preferring other candidates nearly 100% of the time. www.geekwire.com/2024/ai-over...
geekwire.com
AI overwhelmingly prefers white and male job candidates in new test of resume-screening bias
Kyra Wilson, left, University of Washington doctoral student in the Information School and Aylin Caliskan, a UW assistant professor in the iSchool. (UW
013