How AI Deepfakes and Bias Threaten Elections, Health Care and Public Trust - TCU Magazine TCU researchers expose how AI deepfakes threaten elections, health care and public trust through misinformation and bias.

It took 18 years here, but I finally made it into the annual TCU research magazine! This nice article features some of the AI regulation work I've done with @jeremylittau.com and @shoreingber.bsky.social over the past couple of years. magazine.tcu.edu/endeavors-20...

“AI may be everywhere, but it’s not for everyone - at least not yet. If we continue to ignore the perspectives of marginalized people, we risk building an AI-powered future that deepens inequities rather than reducing them.”

Turns out many marginalized Americans are skeptical of / resistant to AI. Thanks Jared Wadley at @umichnews.bsky.social for covering our AI attitudes research! Co-authors: @naz-andalibi.bsky.social, @mayworms.info, @shoreingber.bsky.social

news.umich.edu/marginalized...

dl.acm.org/doi/10.1145/...

Alexis Shore Ingber presenting at the FAccT 2025 conference. Slide lists key takeaways: 1. People have significantly more negative attitudes toward emotion AI than AI. 2. People view all input data to emotion AI as sensitive. 3. Many fear negative consequences if their emotion data is shared, especially with powerful actors.

@shoreingber.bsky.social presents our work with @naz-andalibi.bsky.social about people’s perceptions of different types of emotion AI and potential consequences #FAccT2025 dl.acm.org/doi/pdf/10.1...

Finally, I was on the team for two more wonderful FAccT papers!:

1) Technical Solutions to Emotion AI's Privacy Harms: A Systematic Literature Review

2) AI Attitudes Among Marginalized Populations in the U.S.: Nonbinary, Transgender, and Disabled Individuals Report More Negative AI Attitudes

Links to the papers:
Emotion AI Regulation: tinyurl.com/bdf7nfth
Emotion AI in Hiring: tinyurl.com/avabvj6d
Sensitivity of Emotion AI Input Data & Recipients: tinyurl.com/3x879x3x

On Tuesday (11am) I'll present 2 more papers focused on emotion AI: 1) an experiment comparing job-seekers engagement with a (fake) emotion AI hiring software with non-emotion AI/human and 2) a survey gauging sensitivity perceptions around emotion AI input data and recipients
@haimson.bsky.social

Thrilled to be at #FAccT2025 sharing the work
@naz-andalibi.bsky.social and I have been developing over the past year! 🥹 Tomorrow, I'll be presenting a paper which provides empirical evidence supporting regulation of emotion AI across contexts of deployment (11:45 AM)

Finally, I was on the team for two more wonderful FAccT papers!:

1) Technical Solutions to Emotion AI's Privacy Harms: A Systematic Literature Review

2) AI Attitudes Among Marginalized Populations in the U.S.: Nonbinary, Transgender, and Disabled Individuals Report More Negative AI Attitudes

Links to the papers:

Emotion AI Regulation: tinyurl.com/bdf7nfth
Emotion AI in Hiring: tinyurl.com/avabvj6d
Sensitivity of Emotion AI Input Data & Recipients: tinyurl.com/3x879x3x

On Tuesday (11am) I'll present 2 more papers focused on emotion AI: 1) an experiment comparing job-seekers engagement with a (fake) emotion AI hiring software with non-emotion AI/human and 2) a survey gauging sensitivity perceptions around emotion AI input data and recipients @haimson.bsky.social

Excited to share our latest article (with the amazing @jonpenny.bsky.social and @shoreingber.bsky.social) “The Chilling Effects of Dobbs” in Florida Law Review. The piece grew out of Jon’s creative study and our broader work together on intimate privacy. Here www.floridalawreview.com/article/1381...

I participated in the Body politics workshop on Saturday. Shoutout to organizers and participants! And today, I will present work done with @shoreingber.bsky.social about #emotionAI across domains in the US. 2:46pm, room G302 #CHI2025

I’m the Canadian who was detained by Ice for two weeks. It felt like I had been kidnapped I was stuck in a freezing cell without explanation despite eventually having lawyers and media attention. Yet, compared with others, I was lucky

U.S.: insurrectionists go free, but women who've committed no crimes are "bound in chains that wrapped tightly around our waists, with our cuffed hands secured to our bodies and shackles restraining our feet, forcing every movement into a slow, clinking struggle." www.theguardian.com/us-news/2025...

Happy to share a #CHI25 paper with Jim Cummings & Danny Jia 🎉 We demonstrate that in social VR, 1) interruptions from unknown users reduce disclosure and 2) seeing a conversation partner’s offline profile—mediated by the belief that they are a real person—encourages disclosure. tinyurl.com/35y2dd5m

3) individuals with disabilities and minoritized genders were
significantly less comfortable than others across a variety of contexts; and 4) perceived accuracy explained a large proportion of the variance in comfort levels across contexts. #CHI2025

1) Although comfort was distinct across 11 contexts, even the most favorable context (healthcare) yielded low comfort levels; 2) participants were significantly more comfortable with inferences of happiness and surprise compared to other emotions; #CHI2025

Thrilled to share #CHI2025 paper with @shoreingber.bsky.social where we look at US public's attitudes towards emotion AI in general and in 11 contexts including healthcare, work, employment, border control, cars, kids' toys, social media, education, public spaces, etc. Preprint: shorturl.at/sgOiA

Elon Musk’s DOGE is feeding sensitive federal data into AI to target cuts Department of Government Efficiency workers embedded in the Education Department have been processing sensitive agency data using artificial intelligence.

I spoke to @washingtonpost.com about reports that AI tools are being used with federal data noting that this is occurring without transparency, consent, or attention to privacy and security concerns, and heightening the American public's mistrust of AI www.washingtonpost.com/nation/2025/...

Bias in Technology: The Causes, Consequences, and Possible Solutions Panelists will discuss how bias affects the technology that has become pervasive in our lives -- and efforts to address it.

I will be speaking at UNH Franklin Pierce School of Law next Monday at 5PM EST as part of a panel discussion about AI bias, offering prospective solutions for law and policy. For more information and to register (there is a Zoom option!) visit here: law.unh.edu/blog/2025/01...

This International Privacy Day, Choosing Hope Over Despair Fighting for privacy rights has always been an uphill battle, perhaps now more than ever. Despite periodic moments of public outcry over abuses of state and corporate power enabled by egregious data p...

Today is International Privacy Day. In light of the systematic dismantling of our federal government, coordinated assaults on society’s most vulnerable, and widespread craven capitulation—why should anyone care about privacy right now? 🧵

cdt.org/insights/thi...

This is obviously extremely bad; I'll also just note that they're kneecapping the actual engine of U.S. innovation--our funding of university led basic research-- at the same time China shows how close they are to overtaking US tech. Incredibly stupid.

💡NEW SCREENSHOT RESEARCH! I empirically demonstrate interpersonal and affordance-based interventions that can mitigate screenshot collection and sharing of private messages (+ some privacy theory building 🤓): t.co/74NuQ8izsy

Opinion: The TikTok ban is already dead in the water No matter what happens next, Donald Trump has already completely undermined the law’s efforts to protect national security from Chinese interference

I wrote about the the TikTok ban in the @theglobeandmail.com. I argue the ban is dead for a few reasons, but mostly because its' design failed to account for the new era that Trump's admin represents: persistent weaponization of tech and state power: www.theglobeandmail.com/opinion/arti...