iiinova — where your digital self keeps evolving

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

👋🏽 Happening soon! Come join us at P1 Room 114 at 9am to learn how to make scam education fun AND effective.

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

📜 You can read the full paper (open access) here: dl.acm.org/doi/epdf/10.... If you're at #CHI2026, Owen and Steven will be presenting their paper tomorrow (Wednesday, 4/15). Say hi if you see them around! 📍P1, Room 114 – Conversational AI, Agency and Control ⏰ 9am

000

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

💡Qualitatively, we find that users more frequently provided action-oriented advice over urging caution or providing emotional support. Overall, ScamPilot demonstrates the potential for multi-agent conversational user interfaces to augment learning and improve scam education.

100

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

📊 Through a between-subjects study (N=150), we find that blending advice-giving w/ MCQs significantly increased scam recognition (+8%) without decreasing wariness towards legitimate conversations. Users’ response efficacy and change in self-efficacy was also 9% and 19% higher, respectively.

100

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

In ScamPilot users are presented with a real-time scam simulation with two large language models (LLMs): a *scammer* agent and a *target* agent. Users take on the role of a *helpful advisor* to support the target and prevent them from falling for the scam.

100

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

👨‍🔬 In our #CHI2026 paper, undergrads Owen Hoffman, Steven Peng, Sajid Kamal, and Alex You built and evaluated ScamPilot to advance the state of the art in scam education.

110

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

📚 Giving people information about scams — through popups, warnings, text and video messages — isn't enough. Scammers also constantly adapt their tactics to bypass automated detection methods. Training modules can be monotonous and ineffective. So what can we do? [ tinyurl.com/scam-trainin... ]

100

Sukrit Venkatagiri @sukritv.bsky.social · 15/04/2026

🚨 Have you or someone you know been impacted by a #scam? You're not alone. 73% of US adults have been affected by scams in the last year, leading to over $16 billion reported lost. Imposter scams alone cost Americans $3 billion. [ www.axios.com/2025/04/23/fbi-internet-crime-loss-record-high-2024 ]

100

Sukrit Venkatagiri @sukritv.bsky.social · 21/10/2025

🔎 Do you work at a public university or collaborate w/ someone who does? Come learn what happens when your messages are requested through public records laws in my #CSCW2025 paper w/ Rachel, @emmaspiro.bsky.social, & @katestarbird.bsky.social 🕝 2:30pm today 📍Bekken, Core Concepts in Privacy Research

1132

Sukrit Venkatagiri @sukritv.bsky.social · 14/10/2023

Ever notice how reposts are used differently vs. replies? In our ACM COMPASS 2023 paper we found that replies *reframed* the conversation of a post compared to quote retweets on Twitter. But retweets from *politically opposed* accounts twisted the OPs' words. dl.acm.org/doi/10.1145/...

120

Sukrit Venkatagiri @sukritv.bsky.social · 05/09/2023

First day teaching Social Computing at Swarthmore College! We did a speculative design exercise — imagining what social technologies will be like in 10 yrs. Students also grappled w/ how we can better design the systems that we inhabit, why we have a responsibility to do so, & why it's so difficult.

Barbie dancing asking "Do you guys ever think about social computing" (used to say "dying." Title is Social and Crowd Computing, CPSC91R, Fall 2023. Dr. Sukrit Venkatagiri.

Barbie standing off her porch, viewing a pink barbie world/city. Overlaid is text "social computing" and the question "how can we design the social systems that we inhabit?"

White background with text overlaid: "About Me" and "I am Kenough" below it (quote from the Barbie movie).

391

Sukrit Venkatagiri @sukritv.bsky.social · 11/08/2023

I had a wonderful year as a postdoc at UW working with @katestarbird.bsky.social & so many amazing students. Now, I'm excited to start as an assistant professor of computer science at Swarthmore College where I'll direct the Collective Resilience Lab. Here's a few photos from my first week!

1141

Sukrit Venkatagiri @sukritv.bsky.social · 09/06/2023

Is there political bias in crowdsourced content #moderation? In our #FAccT2023 paper led by Jacob Thebault-Spieker w/ Naomi Mine & Kurt Luther we find risks of political bias but aggregation techniques that account for diverse political perspectives can mitigate it. arxiv.org/abs/2305.14500

Screenshot of paper title, authors, abstract, and ACM reference format. "Diverse Perspectives Can Mitigate Political Bias in Crowdsourced Content Moderation" by Jacob Thebault-Spieker, Sukrit Venkatagiri, Naomi Mine, and Kurt Luther.

030

Sukrit Venkatagiri @sukritv.bsky.social · 23/05/2023

After the January 6th attack, an online community of sleuths formed on Twitter, called the Sedition Hunters. They helped the FBI+DOJ identify 100s of suspected rioters. Check out our #TheWebConf2023 paper where we take a closer look at the community! dl.acm.org/doi/10.1145/3543507.3583…

130

Sukrit Venkatagiri @sukritv.bsky.social · 10/05/2023

I guess this photo is an appropriate first skeet. Hello, Bluesky.

Photo of the Washington Monument in front of the reflecting pool on a sunny day.

020