Way back when, I wrote a PhD dissertation on design patterns for improving this sort of UI. One key recommendation was to make the safer/recommended option more prominent after I did multiple studies showing how most people instinctively swat that without reading.

I said an exasperated "oof" out loud after reading the end of this great @kashhill.bsky.social & @jenvalentino.bsky.social piece. www.nytimes.com/2025/11/23/t...

The most thorough reporting I’ve seen on how Chat GPT is affecting some people’s mental health, (and what the company is doing about it). Great work from @kashhill.bsky.social and @Jen Valentino www.nytimes.com/2025/11/23/t...

The company essentially turned a dial that made ChatGPT more appealing and made people use it more, but sent some of them into delusional spirals. OpenAI has since made the chatbot safer, but that comes with a tradeoff: less usage.

Some stand-out reporting by my colleague @kashhill.bsky.social on OpenAI's tightroping this year between engaging more ChatGPT users and making them lose touch with reality www.nytimes.com/video/techno...

For the last few months, we've been talking to current and former employees of OpenAI to understand what went wrong with ChatGPT this year and how the company is fixing it. Here's the story: www.nytimes.com/2025/11/23/t...

For everyone writing yearend stories about AI lovers, don't make the mistake of thinking the number of weekly visitors to the /MyBoyfriendIsAI subreddit (which is what gets displayed now) is the number of members. Have seen two major pubs now make this mistake. And I get it! I did too (on Bluesky).

Allan Brooks, the corporate recruiter from Canada I wrote about in August who went into a 3-week-long delusional spiral with ChatGPT, sued OpenAI Thursday, alongside six other plaintiffs. They blame ChatGPT for their mental breakdowns and for four suicides. www.nytimes.com/2025/11/06/t...

Character.AI has resources for kids who are about to lose access to their chatbots. One is essentially “consider reading a book.” support.character.ai/hc/en-us/art...

I read this while on a 10 hour road trip with my kids. I used to do these road trips when I was a kid and I would read books the whole time. My kids did read books but they spent more time watching an iPad. Felt more guilty about that than normal thanks to this: thebaffler.com/salvos/we-us...

Character.AI plans to stop offering chatbots to users under 18 www.nytimes.com/2025/10/29/t...

I’ll be in Cambridge on Thursday talking about chatbots. Free event if you’re interested: cyber.harvard.edu/events/frien...

Really struck by how brilliant a protesting tactic this is not just in terms of optics, but as a form of both privacy protection and discouragement of violence. It masks your face and has to give pause to anyone thinking about beating you up.

alive internet theory

Today my @nytimes.com colleagues and I are launching a new series called Lost Science. We interview US scientists who can no longer discover something new about our world, thanks to this year‘s cuts. Here is my first interview with a scientist who studied bees and fires. Gift link: nyti.ms/3IWXbiE

What strikes me about this is how much more clearly incriminating chats with ChatGPT are going to be than Google searches in criminal investigations

This is like a tweet thread from back when social media was good.

This is absolutely wild. I’m on a cross country flight. We are being diverted midway through to Denver. The reason? Some dude is sitting in the exit row who didn’t pay the $155 fee and he refuses to move back to his seat.

I just finished reading @kashhill.bsky.social 'Your Face Belongs to Us' and it's absolutely fucking tremendous and fucking terrifying in equal measure. If you care about privacy and technology it's a must read. Here's the link which has blurbs that more eloquently praise it than I'll be able to!

Outstanding questions from the last few weeks of news: 1. What happened to the $50,000 in the Cava bag? 2. Who were the 17 people in the boats who were killed? 3. Is it safe to fly when the government is shut down?

A way to distance yourself from the bots -- from the NYT comments section. (www.nytimes.com/shared/comme...) For the companies making similar decisions about how the models should act, it's a tradeoff between fun and safety.

Not the point of this piece exactly but a great example of how chatbot validation could increase polarization www.nytimes.com/2025/09/26/w...

A month after my last skeet, the subreddit "My Boyfriend is AI" now has 88,000 members and is the subject of an MIT study that found that "AI companionship emerges unintentionally through functional use rather than deliberate seeking." arxiv.org/html/2509.11...

This is from a story about the president getting rid of federal prosecutors who refuse to follow his marching orders, but this speaks to what seems to be the mindset in general this time around. www.nytimes.com/2025/09/20/u...

Use ChatGPT for relationship advice at your peril.

Thank you! I appreciate you. (If you know, you know.)

OpenAI has a new paper on how people use ChatGPT. I can't help but look at the categories and wonder which ones the various delusional spirals I've reported on slot into.

An “independently published” biography of Charlie Kirk was one of the top selling books on Amazon this week, reaching #12 according to one news report. It was #25 when I looked last night. Reviewers said it was AI-generated nonsense. As of this morning, Amazon has pulled it.

When ChatGPT first took off this was the use case that confused me the most. People said they “only” used it to write “first drafts.” That’s the most important draft! That’s the thinking part!

After FTC launched an inquiry into AI chatbots, particularly dangers toward children, & Senate Judiciary subcommittee holding a hearing today on chatbot harms, OpenAI announced it's going to try to detect which users are under 18 to give them a different experience: openai.com/index/buildi...

I'm on The Daily today talking about my reporting on AI chatbots and how heavy users of ChatGPT can go into delusional, and sometimes tragic, spirals: www.nytimes.com/2025/09/16/p... Also today, an OpenAI blog post from Sam Altman on "safety, freedom & privacy": openai.com/index/teen-s...

"Trapped in a ChatGPT Spiral." 🌀 Important work by @kashhill.bsky.social & @dylanfreedman.nytimes.com on how chatbots have a tendency to endorse conspiratorial and mystical belief systems. This shows again how conformist LLMs can be. Worth a listen 👇 www.nytimes.com/2025/09/16/p...

The Federal Trade Commission is investigating generative A.I. chatbots: www.ftc.gov/news-events/...

Condemn political violence every time it happens. Do not incite it. Do not celebrate it. Do not call for more of it. www.nytimes.com/2025/09/10/o...

Department of Homeland Security re-ups with Clearview AI in what I think is the company's biggest gov deal so far: www.404media.co/ice-spends-m...

And now this www.nytimes.com/2025/09/10/u...

The continuing "this is just how it is and we're not going to explain why" from the Supreme Court is deeply unsettling www.nytimes.com/live/2025/09...

Authors whose books were used in LLM training by Anthropic to get $3,000 each. Adds up to a very big number. Gift link: www.nytimes.com/2025/09/05/t...

More on the quest to turn everything and everyone into a surveillance camera

This story about a venture capitalist who made it her mission to go after her surrogate after a miscarriage is horrifying. It sounds like the community of "intended parents" with whom she was sharing what was happening validated her throughout. And maybe ChatGPT too. www.wired.com/story/the-ba...

What makes me skeptical AGI is around the corner is how ChatGPT is now handling prompts that elicit any kind of self-harm info. Gives a hotline warning, even when the request is a summary of Romeo and Juliet: www.linkedin.com/feed/update/... Reddit is flipping out: www.reddit.com/r/ChatGPT/se...

My story on the safeguards OpenAI plans to add for teens and other users who appear to be in distress. It includes parental controls which its developer community has been requesting since Jan 2024 and routing "sensitive" prompts to GPT-5 thinking. www.nytimes.com/2025/09/02/t...

Targeted killings were one of the most controversial legacies of the Obama administration with drone strikes overseas. The practice is now moving closer to home and the targets are suspected drug dealers. www.nytimes.com/2025/09/02/u... Trump posted the video here: truthsocial.com/@realDonaldT...

"available soon" This should be regulated as a standard practice, before deployment, for all platforms, including AI tools, models, and systems.

Great story by my colleagues on the constant political realignment of Grok, the generative AI chatbot over on the other site. This change to Grok’s answer about the greatest threat to Western civilization is particularly ironic/meta. Gift link: www.nytimes.com/2025/09/02/t...

OpenAI also announced today that it is routing “sensitive” conversations to a version of its chatbot that it says is better at staying on the rails. openai.com/index/buildi...

OpenAI announced today that parental controls will be available soon for minors’ ChatGPT accounts. openai.com/index/buildi...

Butlerian Jihad now: defector.com/butlerian-ji...

A man who killed his mother and himself had been discussing his paranoid thoughts with ChatGPT. This is quite long and I’m very burned out. Does it say how the chats surfaced? www.wsj.com/tech/ai/chat...

i find this image so fascinating. it is both a literal example of authoritarianism but also a second-hand reproduction of the aesthetics of other authoritarian states. it's like a simulacra of authoritarianism whose purpose is to attempt to make the simulacra real.