Профиль

More than $70 million worth of Bitcoin was recently stolen in a hack. One of the goals of cryptocurrency is reducing reliance on trusted intermediaries. In practice, though, trust doesn't disappear, it shifts.

A large Meta-backed experiment found that most people were exposed to very little content from known “untrustworthy” sources on Facebook and Instagram. theconversation.com/meta-backed-...

Meta-backed research finds exposure to ‘untrustworthy’ social media is rare. The fine print is less reassuringThere is a risk this study will be leveraged by big tech platforms to exonerate themselves.theconversation.com

12 of the largest right-wing influencers online have collectively seen their engagement plummet over the past year.

“the researchers argue, that AI chatbots could soon take over much of the scam process as fully independent fraud agents” www.wired.com/story/ai-sca...

AI Scammers Are Better at Building Trust Than HumansResearchers pitted a person against a Claude agent and found that, after a week of texting, the AI chatbot was more effective at creating “exploitable trust” with others.www.wired.com

I built four free tools for navigating information online, and a guide for teachers, librarians, and facilitators on how to use them. They walk through four questions: Who am I? Who do I talk to and hear from? What deserves my attention? How do I check this claim?

A new large-scale European study finds that when people feel their identity is under threat, they tend to prioritize security, control, and tradition over openness, and become less supportive of liberal democratic policies. onlinelibrary.wiley.com/doi/10.1111/...

Identity threat, personal values, and political attitudes: Evidence from a large‐scale survey across the European UnionIn political contexts, identities are typically based on shared values. Populist rhetoric often utilizes real and perceived threats to evoke such identities, especially on the far right. Using dedica...onlinelibrary.wiley.com

"Scientists did their darnedest within the systems they were handed and the polarized landscape we find ourselves in." yourlocalepidemiologist.substack.com/p/its-time-t...

It's time to leave the man alone.One day in 2020, I went to a coffee shop.yourlocalepidemiologist.substack.com

A large field experiment found that platforms can dramatically reduce exposure to repeat untrustworthy sources, but doing so did not produce measurable changes in users' beliefs or attitudes. www.science.org/doi/10.1126/...

Untrustworthy sources on Facebook and Instagram in 2020: Concentrated exposure but no attitudinal effectsReducing exposure to untrustworthy sources on Facebook/Instagram had no measurable effects on attitudes, even for heavy consumers.www.science.org

The incentives in American politics make going viral more rewarding than governing well. It's no wonder so much of what passes for political discourse is really just about creating content.

Why do people believe conspiracy theories? Research points to unmet needs for certainty, control, and belonging. My own research suggests that perceived victimhood is another key factor. medium.com/@m.j.faccian...

What Makes Someone Believe a Conspiracy Theory?A massive new meta-analysis concludes that certainty, security, and belonging play a major role, and my own research finds that perceived…medium.com

My Indonesian media literacy game (Gali Fakta) was one of 72 interventions in this review. Solid critique: most studies never actually tested the theory's core mechanisms (threat, counterarguing), just assumed them. journals.sagepub.com/doi/full/10....

Sage Journals: Discover world-class researchSubscription and open access journals from Sage, the world's leading independent academic publisher.journals.sagepub.com

This study interviewed young adults who expressed support for conspiracy theories and found that, for many participants, conspiracy theories provided a sense of identity, community, and agency by helping them make sense of feelings of exclusion and powerlessness.

Why do influencers often get a pass for spreading inaccurate information? This paper argues that influencers cultivate "authentic expertise" by emphasizing personal experience, vulnerability, and transparency, making audiences more likely to forgive inaccurate information.

"The research found that AI-generated content appeared in 40% of top health-related TikTok videos, with most posted by a growing number of deepfake influencer accounts designed to resemble real doctors" www.theguardian.com/technology/2...

Misleading AI-generated doctors pose ‘huge danger to public safety’Research shows AI-generated physicians are gaining millions of views on TikTok by spreading dubious health advicewww.theguardian.com

A new study found that climate misinformation spikes during natural disasters, spreads through coordinated networks, and is amplified across low-moderation platforms such as Telegram. misinforeview.hks.harvard.edu/article/ampl...

Amplifying the storm: Climate disinformation dynamics during natural disasters on right-wing extremist Telegram channels | HKS Misinformation ReviewClimate change amplifies natural disasters, posing an existential threat to our society. However, a digital storm is raging on right-wing extremist Telegram channels where climate disinformation works...misinforeview.hks.harvard.edu

Mayor Mamdani advocates for critical ignoring here! youtube.com/shorts/X3xz8...

Feedback vs. Noise | The Weekly Show with Jon StewartYouTube video by The Weekly Show with Jon Stewartyoutube.com

Two new episodes of Misguided: The Audio Edition are live! Chapter 1 covers the scope and consequences of misinformation & Chapter 2 goes inside the psychology of identity. Both have updated commentary with my thoughts a year after the book came out. matthewfacciani.substack.com/p/misguided-...

Misguided: The Audio Edition, Chapters 1 and 2Two new audio chapters: the scope and consequences of misinformation, and the identity machinery that makes us more likely to believe it.matthewfacciani.substack.com

Improving science literacy requires teaching people how science actually works: that it's fallible, self-correcting, and how it's funded. This paper argues that strategy is more helpful rather than just teaching more scientific facts. journals.sagepub.com/doi/full/10....

Trump's teleprompter operator allegedly made ~$100,000 betting on his own boss's speeches on Kalshi. Wild story on its own. But it makes me wonder: how much larger insider trading, by people with even better access, is never getting caught at all? abcnews.com/US/white-hou...

White House teleprompter operator made more than $100K betting on Trump's speeches: SourcesPresident Trump's longtime teleprompter operator is believed to have made tens of thousands of dollars by placing bets on the contents of Trump's speeches, sources say.abcnews.com

Here is a quick test of Pangram, the AI Detection tool that Substack will be using to evaluate AI writing on the platform. I used an AI agent to generate the first 200 words of text, and then I rambled on myself with my human brain after. It concluded the entire block of text was 100% human.