Matt DeVerna
@matthewdeverna.com
Postdoc with Stanford's Tech Impact and Policy Center
(@techimpactpolicy.bsky.social). Formerly IU / Observatory on Social Media.
Computational social science, human-AI interaction, social media, trust and safety, etc.
🧨 matthewdeverna.com
(@techimpactpolicy.bsky.social). Formerly IU / Observatory on Social Media.
Computational social science, human-AI interaction, social media, trust and safety, etc.
🧨 matthewdeverna.com
Pinned
🚨 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
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
Reposted by Matt DeVerna
Is Joe Rogan really just a voice of the right? Our new @csmapnyu.org piece for @goodauth.bsky.social shows he’s just as much a space for the left and the center, too. A look inside today’s surprisingly complicated podcast information ecosystem. 🎙️
goodauthority.org/news/podcast...
goodauthority.org/news/podcast...
December 17, 2025 at 4:39 PM
Is Joe Rogan really just a voice of the right? Our new @csmapnyu.org piece for @goodauth.bsky.social shows he’s just as much a space for the left and the center, too. A look inside today’s surprisingly complicated podcast information ecosystem. 🎙️
goodauthority.org/news/podcast...
goodauthority.org/news/podcast...
❤️❤️❤️❤️❤️
2026 tutorial submissions are open!
✔️ Submission deadline: February 2nd, 2026
✔️Acceptance Notification: February 16th, 2026
View the guidelines here:
lnkd.in/eBHVe2tP
✔️ Submission deadline: February 2nd, 2026
✔️Acceptance Notification: February 16th, 2026
View the guidelines here:
lnkd.in/eBHVe2tP
December 15, 2025 at 3:05 PM
❤️❤️❤️❤️❤️
Reposted by Matt DeVerna
Started a thread in the other place and bringing it over here - I really think we should be more vocal about the opportunities that lay at the intersection of these two options!
So I'm starting a live thread of new roles as I become aware of them - feel free to add / extend / share :
So I'm starting a live thread of new roles as I become aware of them - feel free to add / extend / share :
Life situations are bleak right now for a lot of people. In tech, the "Venn Diagram" of (1) positive work and (2) making enough money to support your family is increasingly non-overlapping. We all do what we can.
This image has been living in my mind rent-free for months.
This image has been living in my mind rent-free for months.
October 29, 2025 at 2:25 AM
Started a thread in the other place and bringing it over here - I really think we should be more vocal about the opportunities that lay at the intersection of these two options!
So I'm starting a live thread of new roles as I become aware of them - feel free to add / extend / share :
So I'm starting a live thread of new roles as I become aware of them - feel free to add / extend / share :
The incredible @yyahn.bsky.social is building something special at UVA.
🚨 If you're a fit for this job, I highly recommend applying!
Find more details via the link below.
www.linkedin.com/posts/yyahn_...
🚨 If you're a fit for this job, I highly recommend applying!
Find more details via the link below.
www.linkedin.com/posts/yyahn_...
December 10, 2025 at 6:01 PM
The incredible @yyahn.bsky.social is building something special at UVA.
🚨 If you're a fit for this job, I highly recommend applying!
Find more details via the link below.
www.linkedin.com/posts/yyahn_...
🚨 If you're a fit for this job, I highly recommend applying!
Find more details via the link below.
www.linkedin.com/posts/yyahn_...
Reposted by Matt DeVerna
Keynote Announcement!
We're thrilled to have Kate Starbird from the University of Washington's Department of Human Centered Design & Engineering speak at IC2S2 2026.
Registration is open with submissions opening on 12/15: ic2s2-2026.org
We're thrilled to have Kate Starbird from the University of Washington's Department of Human Centered Design & Engineering speak at IC2S2 2026.
Registration is open with submissions opening on 12/15: ic2s2-2026.org
December 8, 2025 at 4:13 PM
Keynote Announcement!
We're thrilled to have Kate Starbird from the University of Washington's Department of Human Centered Design & Engineering speak at IC2S2 2026.
Registration is open with submissions opening on 12/15: ic2s2-2026.org
We're thrilled to have Kate Starbird from the University of Washington's Department of Human Centered Design & Engineering speak at IC2S2 2026.
Registration is open with submissions opening on 12/15: ic2s2-2026.org
Reposted by Matt DeVerna
What if you could see fewer hostile political posts on social media? A new paper out in Science by Martin Saveski @msaveski.bsky.social of the iSchool, along with @tiziano.bsky.social, @jiachenyan.bsky.social, Jeff Hancock, Jeanne Tsai and @mbernst.bsky.social, explores this: doi.org/10.1126/scie...
December 4, 2025 at 10:14 PM
What if you could see fewer hostile political posts on social media? A new paper out in Science by Martin Saveski @msaveski.bsky.social of the iSchool, along with @tiziano.bsky.social, @jiachenyan.bsky.social, Jeff Hancock, Jeanne Tsai and @mbernst.bsky.social, explores this: doi.org/10.1126/scie...
Reposted by Matt DeVerna
In what I think may be previously unreported news, the SF City Attorney's office told me that they are confident more AI nudifiers will be shutting down as a consequence of their lawsuit -- including ones based overseas.
December 4, 2025 at 2:22 PM
In what I think may be previously unreported news, the SF City Attorney's office told me that they are confident more AI nudifiers will be shutting down as a consequence of their lawsuit -- including ones based overseas.
Reposted by Matt DeVerna
If platforms won't do it out of the good of their hearts, they may need to be coerced.
California's AB 621, passed in October, may make them responsible for providing services to nonconsensual deepfake nude providers if they're notified and don't act.
Here's what the author of the law told me:
California's AB 621, passed in October, may make them responsible for providing services to nonconsensual deepfake nude providers if they're notified and don't act.
Here's what the author of the law told me:
December 4, 2025 at 2:18 PM
If platforms won't do it out of the good of their hearts, they may need to be coerced.
California's AB 621, passed in October, may make them responsible for providing services to nonconsensual deepfake nude providers if they're notified and don't act.
Here's what the author of the law told me:
California's AB 621, passed in October, may make them responsible for providing services to nonconsensual deepfake nude providers if they're notified and don't act.
Here's what the author of the law told me:
Reposted by Matt DeVerna
The lowest-hanging fruit should be deplatforming.
There is no reason Meta can't get better at blocking ads like these, nor Apple from hosting the underlying apps. There is no reason for Google to provide single-sign on to 5 of the top 10 nudifiers, or Cloudflare to provide CDN services.
There is no reason Meta can't get better at blocking ads like these, nor Apple from hosting the underlying apps. There is no reason for Google to provide single-sign on to 5 of the top 10 nudifiers, or Cloudflare to provide CDN services.
December 4, 2025 at 2:15 PM
The lowest-hanging fruit should be deplatforming.
There is no reason Meta can't get better at blocking ads like these, nor Apple from hosting the underlying apps. There is no reason for Google to provide single-sign on to 5 of the top 10 nudifiers, or Cloudflare to provide CDN services.
There is no reason Meta can't get better at blocking ads like these, nor Apple from hosting the underlying apps. There is no reason for Google to provide single-sign on to 5 of the top 10 nudifiers, or Cloudflare to provide CDN services.
Reposted by Matt DeVerna
For today's article, I spoke to a dozen experts working on countermeasures to this problem.
I came out of it with uncharacteristic optimism that the concerted effort of the many people focusing on this problem may start yielding results in 2026. But it’s going to take work.
I came out of it with uncharacteristic optimism that the concerted effort of the many people focusing on this problem may start yielding results in 2026. But it’s going to take work.
Nonconsensual nude generators had another banner year. What will it take to defeat them?
Deplatforming the companies, debilitating the technology, and deterring the users
indicator.media
December 4, 2025 at 2:13 PM
For today's article, I spoke to a dozen experts working on countermeasures to this problem.
I came out of it with uncharacteristic optimism that the concerted effort of the many people focusing on this problem may start yielding results in 2026. But it’s going to take work.
I came out of it with uncharacteristic optimism that the concerted effort of the many people focusing on this problem may start yielding results in 2026. But it’s going to take work.
Reposted by Matt DeVerna
I’m sick and tired of writing about these awful tools. But if we think about digital safety in terms of reducing harms that are both prevalent and acute, this has to be one of the top current concerns from generative AI. Nudifiers reach millions and make millions.
Nonconsensual nude generators had another banner year. What will it take to defeat them?
Deplatforming the companies, debilitating the technology, and deterring the users
indicator.media
December 4, 2025 at 2:13 PM
I’m sick and tired of writing about these awful tools. But if we think about digital safety in terms of reducing harms that are both prevalent and acute, this has to be one of the top current concerns from generative AI. Nudifiers reach millions and make millions.
Reposted by Matt DeVerna
Today on Indicator: 2025 has been a banner year for AI nudifiers. I found another 9,000 ads on Meta since my last report, bringing the total for this year to 25,000. The top 10 nudifying websites got 10 million views in October.
Nonconsensual nude generators had another banner year. What will it take to defeat them?
Deplatforming the companies, debilitating the technology, and deterring the users
indicator.media
December 4, 2025 at 2:12 PM
Today on Indicator: 2025 has been a banner year for AI nudifiers. I found another 9,000 ads on Meta since my last report, bringing the total for this year to 25,000. The top 10 nudifying websites got 10 million views in October.
Reposted by Matt DeVerna
Happy to share our most recent paper, led by @allisonwan.bsky.social, auditing what domains Google shows when you search for a Member of Congress.
journalqd.org/article/view...
@journalqd.bsky.social
journalqd.org/article/view...
@journalqd.bsky.social
December 4, 2025 at 2:42 AM
Happy to share our most recent paper, led by @allisonwan.bsky.social, auditing what domains Google shows when you search for a Member of Congress.
journalqd.org/article/view...
@journalqd.bsky.social
journalqd.org/article/view...
@journalqd.bsky.social
Reposted by Matt DeVerna
IC2S2 2026 registration is open!
Explore the 2026 conference here ➡️ ic2s2-2026.org
✔️ Submissions open December 15th
✔️ Keynotes will be announced between now and February
✔️ Full program of selected talks and tutorials will be available in late April
Explore the 2026 conference here ➡️ ic2s2-2026.org
✔️ Submissions open December 15th
✔️ Keynotes will be announced between now and February
✔️ Full program of selected talks and tutorials will be available in late April
December 1, 2025 at 9:40 PM
IC2S2 2026 registration is open!
Explore the 2026 conference here ➡️ ic2s2-2026.org
✔️ Submissions open December 15th
✔️ Keynotes will be announced between now and February
✔️ Full program of selected talks and tutorials will be available in late April
Explore the 2026 conference here ➡️ ic2s2-2026.org
✔️ Submissions open December 15th
✔️ Keynotes will be announced between now and February
✔️ Full program of selected talks and tutorials will be available in late April
Reposted by Matt DeVerna
I am pleased to announce that our work “A longitudinal analysis of misinformation, polarization, and toxicity on Bluesky after its public launch” has been accepted at #OSNEM.
The paper is an extension of our previous work presented at #ASONAM.
Don't miss it: www.sciencedirect.com/science/arti...
The paper is an extension of our previous work presented at #ASONAM.
Don't miss it: www.sciencedirect.com/science/arti...
A longitudinal analysis of misinformation, polarization and toxicity on Bluesky after its public launch
Bluesky is a decentralized, Twitter-like social media platform that has rapidly gained popularity. Following an invite-only phase, it officially opene…
www.sciencedirect.com
December 1, 2025 at 12:14 PM
I am pleased to announce that our work “A longitudinal analysis of misinformation, polarization, and toxicity on Bluesky after its public launch” has been accepted at #OSNEM.
The paper is an extension of our previous work presented at #ASONAM.
Don't miss it: www.sciencedirect.com/science/arti...
The paper is an extension of our previous work presented at #ASONAM.
Don't miss it: www.sciencedirect.com/science/arti...
🚨 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
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
November 29, 2025 at 10:06 PM
🚨 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
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
Reposted by Matt DeVerna
New study finds that down-ranking hostile political content in people’s social media feeds decreases political polarization.
www.science.org/doi/10.1126/...
www.science.org/doi/10.1126/...
Reranking partisan animosity in algorithmic social media feeds alters affective polarization
Today, social media platforms hold the sole power to study the effects of feed-ranking algorithms. We developed a platform-independent method that reranks participants’ feeds in real time and used thi...
www.science.org
November 28, 2025 at 11:16 PM
New study finds that down-ranking hostile political content in people’s social media feeds decreases political polarization.
www.science.org/doi/10.1126/...
www.science.org/doi/10.1126/...
Reposted by Matt DeVerna
The UW Center for an Informed Public is looking for postdocs (for 2026-2028) from across diverse disciplines whose research sheds light on the challenges of our modern information environment, promotes civic health, and/or helps people/communities navigate online spaces: apply.interfolio.com/177901
Apply - Interfolio
{{$ctrl.$state.data.pageTitle}} - Apply - Interfolio
apply.interfolio.com
November 24, 2025 at 6:26 PM
The UW Center for an Informed Public is looking for postdocs (for 2026-2028) from across diverse disciplines whose research sheds light on the challenges of our modern information environment, promotes civic health, and/or helps people/communities navigate online spaces: apply.interfolio.com/177901
Reposted by Matt DeVerna
One week left! Submit your paper for the Journal of Online Trust & Safety’s Spring 2026 general issue. ⤵️
🗓 Key Dates
➤ Peer-reviewed research articles due: December 1, 2025
➤ Commentaries due: March 1, 2026
➤ Publication date: April 2026
👉 Submit Your Paper
bit.ly/4oi8b97
#JOTS #TrustAndSafety
🗓 Key Dates
➤ Peer-reviewed research articles due: December 1, 2025
➤ Commentaries due: March 1, 2026
➤ Publication date: April 2026
👉 Submit Your Paper
bit.ly/4oi8b97
#JOTS #TrustAndSafety
November 24, 2025 at 5:25 PM
One week left! Submit your paper for the Journal of Online Trust & Safety’s Spring 2026 general issue. ⤵️
🗓 Key Dates
➤ Peer-reviewed research articles due: December 1, 2025
➤ Commentaries due: March 1, 2026
➤ Publication date: April 2026
👉 Submit Your Paper
bit.ly/4oi8b97
#JOTS #TrustAndSafety
🗓 Key Dates
➤ Peer-reviewed research articles due: December 1, 2025
➤ Commentaries due: March 1, 2026
➤ Publication date: April 2026
👉 Submit Your Paper
bit.ly/4oi8b97
#JOTS #TrustAndSafety
November 22, 2025 at 8:45 PM
Reposted by Matt DeVerna
Don't forget our next OSoMe Awesome Speaker this Wednesday!
📅 November 12 @ 12 PM ET
Petter Törnberg, University of Amsterdam
Register at osome.iu.edu/events/speak...
📅 November 12 @ 12 PM ET
Petter Törnberg, University of Amsterdam
Register at osome.iu.edu/events/speak...
November 10, 2025 at 10:40 PM
Don't forget our next OSoMe Awesome Speaker this Wednesday!
📅 November 12 @ 12 PM ET
Petter Törnberg, University of Amsterdam
Register at osome.iu.edu/events/speak...
📅 November 12 @ 12 PM ET
Petter Törnberg, University of Amsterdam
Register at osome.iu.edu/events/speak...
Reposted by Matt DeVerna
Is social media to blame for teens’ mental health decline?
Intuition says yes—but RCTs find only small short-term effects when users quit.
This new preprint argues these studies don’t prove social media isn’t to blame.
Here’s why: 3 reasons.
Intuition says yes—but RCTs find only small short-term effects when users quit.
This new preprint argues these studies don’t prove social media isn’t to blame.
Here’s why: 3 reasons.
November 9, 2025 at 9:16 PM
Is social media to blame for teens’ mental health decline?
Intuition says yes—but RCTs find only small short-term effects when users quit.
This new preprint argues these studies don’t prove social media isn’t to blame.
Here’s why: 3 reasons.
Intuition says yes—but RCTs find only small short-term effects when users quit.
This new preprint argues these studies don’t prove social media isn’t to blame.
Here’s why: 3 reasons.
Reposted by Matt DeVerna
🚨 New paper! 🔍
“Question the Questions: Auditing Representation in Online Deliberative Processes”
In deliberative polls, participants propose questions for experts but only a few make it to the panel. How representative are those chosen questions of everyone’s interests? 🧵👇
“Question the Questions: Auditing Representation in Online Deliberative Processes”
In deliberative polls, participants propose questions for experts but only a few make it to the panel. How representative are those chosen questions of everyone’s interests? 🧵👇
November 7, 2025 at 7:00 PM
🚨 New paper! 🔍
“Question the Questions: Auditing Representation in Online Deliberative Processes”
In deliberative polls, participants propose questions for experts but only a few make it to the panel. How representative are those chosen questions of everyone’s interests? 🧵👇
“Question the Questions: Auditing Representation in Online Deliberative Processes”
In deliberative polls, participants propose questions for experts but only a few make it to the panel. How representative are those chosen questions of everyone’s interests? 🧵👇