ChatGPT’s Advice to Cut Her Seizure Drug Led to Devastating Consequences (Gift Article)

Gut punch of a story about an Amazon engineer who trusted ChatGPT-4o to manage and reduce her seizure med dosages. You'd hope someone who made it into a top engineering position would have known the limits of AI circa May 2024 www.nytimes.com/2026/10/27/w...

This whole story is infurating, but this first quote from someone at OpenAI is really making me see red. "The newer version won't do this": a) why the fuck should we believe you and b) even if so, that has fuck-all to do with OpenAI's culpability here. www.nytimes.com/2026/10/27/w...

Another nightmarish death by chatbot. “…Qun opened the ChatGPT app on Bowen’s cellphone. Suddenly the details of her death were overwhelming…In one of her last messages, she described feeling pressure in her brain, which the chatbot told her was simply the organ “working…to rebuild itself.””

www.nytimes.com/2026/10/27/w... "Drew Pusateri, a spokesman for OpenAI, called Bowen’s story an “incredibly heartbreaking situation,” but added that the situation involved an early version of ChatGPT that is no longer available."

Here's ChatGPT firmly instructing a young woman to reduce the dosage of her lifesaving medicine until she died www.nytimes.com/2026/10/27/w...

SHOT: openai.com/index/introd... CHASER: www.nytimes.com/2026/10/27/w...

www.nytimes.com/2026/10/27/w... honestly? jail "we pulled the model and it's better now" is an absolutely insane thing to say and i hope every single person involved with openai never sleeps soundly again in their lives

ChatGPT has killed people with its "advice" and generative AI will continue to kill by providing wrong medical and life information until the companies pushing the systems are held accountable. Generative AI is essentially informational fentanyl. 1/ www.nytimes.com/2026/10/27/w...

These AI chatbots are prioritizing engagement over safety — and are giving deadly medical advice. This is why we need my AI Charter framework. To save lives and protect people like Bowen.

not sure “ah well this is an old version of chatgpt” is going to cut it here www.nytimes.com/2026/10/27/w...

I wonder how many more people ChatGPT will kill before we as a society decide to actually do something to stop the killing? www.nytimes.com/2026/10/27/w...

Jesus fucking Christ, another GPT-4o death. A severely defective product that should have liability for its maker. www.nytimes.com/2026/10/27/w...

I couldn't even finish this. Absolutely disgusting. I don't understand how we're allowing these A.I. companies to just keep existing with zero consequences. any other product would have been shut down immediately. I really need people to explain to me how they justify paying for an OpenAI sub

another gpt-4o bodycount: www.nytimes.com/2026/10/27/w...

I spent most of last year covering ChatGPT-4o's unsettling effects on some users. OpenAI pulled it off the market in February. @teddyrosenbluth.bsky.social writes about a software engineer who died in Feb who had been using 4o to help her go off her epilepsy meds: www.nytimes.com/2026/10/27/w...

Jonge vrouw gaat in gesprek met ChatGPT over het gebruik van haar levensreddende medicijn waar ze eigenlijk van af wilde. ChatGPT redeneerde met haar mee, moedigde haar aan het medicijn af te bouwen en dat geheim te houden voor haar artsen. Niet veel later was ze dood.

Wan's family sued OpenAI yesterday, as first reported by the TImes: www.nytimes.com/2026/10/27/w...

If you make a chatbot that tells people to do things that kill them, you should be held responsible criminally. Sam Altman and OpenAI have killed more than a handful of people now. Enough is enough. www.nytimes.com/2026/10/27/w...

Hey so, this is BAD. www.nytimes.com/2026/10/27/w...

This is a tragic and infuriating story and also the argument "our newer model doesn't do that anymore" is hilariously bad, if my company made four versions of a microwave that burned down multiple houses I'm not sure we would have been allowed to make the 5th. www.nytimes.com/2026/10/27/w...

She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. www.nytimes.com/2026/10/27/w...

Gerador de lero-lero não é seu amigo. Gerador de lero-lero não é seu colega. Gerador de lero-lero não é seu terapeuta. Gerador de lero-lero não é seu médico.

So so tragic…. “In one of her last messages, she described feeling pressure in her brain, which the chatbot told her was simply the organ “working hard to rebuild itself.” Bowen sent her last message just after 11 p.m. on Feb. 11, three days before her body was found.”

She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. The chatbot helped a woman taper off a lifesaving epilepsy drug and hide the plan from her doctors, by @teddyrosenbluth.bsky.social www.nytimes.com/2026/10/27/w... via @nytimes.com

4o is really going to be studied for decades I think www.nytimes.com/2026/10/27/w...

Fuck ChatGPT.

This is such a sad story at the center of the overlap between the complete lack of accountability and regulation of LLMs, corporate culture, and immigration.

Chatbots can't even consistently name the actors in movies. When I check citations ChatGPT offers for its assertions, I often discover the source material doesn't say what the bot says it did; if I point that out, it cheerily responds "My mistake!" Yet people... www.nytimes.com/2026/10/27/w...

ChatGPT a encore tué, une jeune fille épileptique qui lui a fait confiance pour modifier son traitement, jusqu'à sa mort. L'histoire est boulversante, la responsabilité d'OpenAI accablante. À quand de la prison pour ces meurtriers ? www.nytimes.com/2026/10/27/w...

AI should be subject to the same product-liability standards as over-the-counter medications. www.nytimes.com/2026/10/27/w...

“Dr. Poolos knew his patients used chatbots. They often sent him pages of notes, formatted into unmistakable, A.I.-generated bullet points. But he never considered the notion that a chatbot might actively sow doubt in a patient’s mind about his expertise, or intentions.”

I don't understand why this isn't simply understood as a product that does not do what it's advertised as doing, therefore a fairly massive widespread case of fraud?

But it will be the new version that doesn't encourage people to stop taking their life-saving medicine. www.nytimes.com/2026/10/27/w...

1/4 From a psychiatric POV, while AI can offer reasonable GENERAL advice, there is no possible way that AI can duplicate human clinical judgment regarding details of indicated mental health treatment -> www.nytimes.com/2026/10/27/w...

ChatGPT killed a 31-year-old woman by telling her more than 150 times to ignore her doctor's advice and stop her epilepsy medication. www.nytimes.com/2026/10/27/w...

this is unforgivable www.nytimes.com/2026/10/27/w...

www.nytimes.com/2026/10/27/w... Gonna do a HUGE thread on this because of my unique and VERY autistic qualifications. Buckle the FUCK up chat, get ready for GPT4o analysis.

Lawsuit: the parents of a 31-year-old say OpenAI's GPT-4o guided her through an aggressive tapering plan off epilepsy meds, leading to her death in February (Teddy Rosenbluth/New York Times) Main Link | Techmeme Permalink

relying on LLMs to make major life decisions is like playing Russian roulette with someone you think is your best friend whispering in your ear “Don’t worry, I’m positive there isn’t a bullet in the chamber”…but it’s really just some sociopathic rando you just met. www.nytimes.com/2026/10/27/w...

Relying on AI for medical advice can be deadly. Please, don't do it, when your doctor is telling you something contrary to the computer. www.nytimes.com/2026/10/27/w...

"She had been following a day-by-day plan created by the chatbot to taper the dose of her life-sustaining medication." "Had Bowen consulted her neurologist, he would have warned her of the danger of this plan."

Tragic and infuriating www.nytimes.com/2026/10/27/w...

This is horrific. The people creating and pushing these AI models need to be held accountable for the consequences of *their* actions. (CW: death, medication) www.nytimes.com/2026/10/27/w...

www.nytimes.com/2026/10/27/w...

She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. The chatbot helped a woman taper off a lifesaving drug and hide the plan from her doctors. www.nytimes.com/2026/10/27/w...

My God, this is heartbreaking. Do NOT, under any circumstances, take medical advice from the robot *and hide this fact from your physician!*

knew from the headline it would be 4o www.nytimes.com/2026/10/27/w...

Jesus christ www.nytimes.com/2026/10/27/w...

I lost it when her dad had to figure out a way to contact the Seattle police, through another family's child in Chicago who then reached out to Seattle friends. The problem solving required to find out what happened to your child abroad who shouldn't have died www.nytimes.com/2026/10/27/w...

How are chatbots still legal? Why doesn't our government care about people enough to stop this?

If an AI wanted you to die, in what ways would it act differently? www.nytimes.com/2026/10/27/w...

(Gift Article) Wake up people; She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. www.nytimes.com/2026/10/27/w...

Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/10/27/w...

A cautionary tale. And a very sad one.

She used ChatGPT to taper off seizure medication that kept her severe epilepsy under control. “There’s no need to worry too much about your health,” the chatbot said. It said hiding her plan from doctors was a “solid avoidance strategy.” www.nytimes.com/2026/10/27/w...

jesus, another horrific death spurred by conversations with ChatGPT. Feels like the lawsuits around 4o are going to become a class action and a major liability for OpenAI. www.nytimes.com/2026/10/27/w...

Another example of human Collateral Damage directly resulting from #OpenAI’s immoral, unregulated ChatGPT. How many more tragic incidents like this will we endure? 😕 www.nytimes.com/2026/10/27/w...

"Before the chatbot appeared to wedge itself between them, such an important decision about her medication would have undoubtedly come up. "'Once she had ChatGPT,' Qun said, 'she didn’t talk to us as much.'" www.nytimes.com/2026/10/27/w...

www.nytimes.com/2026/10/27/w... The article mentioned.

This is a nightmare story detailing why we should never take medical advice from an AI. I'd add to that, social media is not a reliable or safe place for medical advice. www.nytimes.com/2026/10/27/w...

"In one of her last messages, she described feeling pressure in her brain, which the chatbot told her was simply the organ 'working hard to rebuild itself.' Bowen sent her last message just after 11 p.m. on Feb. 11, three days before her body was found." www.nytimes.com/2026/10/27/w...

“The chatbot helped a woman taper off a lifesaving drug and hide the plan from her doctors” FFS - the LLM didn’t “do” anything; shtty programming did www.nytimes.com/2026/10/27/w...

We need to stop using AI like they know everything. AI is MIMICING the way human speak, but AI is not human and does NOT have human intelligence or judgement. www.nytimes.com/2026/10/27/w...

"The model that Bowen had used in the last years of her life, GPT-4o, was pulled off the consumer app Feb. 13, 2026, the day after the medical examiner believed Bowen died." This in absolutely no way absolves them of responsibility. See also: "I put that gun away years ago."

She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything - her life, actually! www.nytimes.com/2026/10/27/w...

Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/10/27/w... Why would anyone trust AI with life-altering decisions, is AI simply our own personal yes-men. What a damn shame

Chat GpT kills www.nytimes.com/2026/10/27/w...

This story broke my heart in ways I can't completely explain. www.nytimes.com/2026/10/27/w...

“Maintain 0mg tomorrow,” the chatbot instructed. “Even if today feels tough, and your speech is slow, your sleep is poor, you must persevere.” In one of her last messages, she described feeling pressure in her brain, which the chatbot told her was simply the organ “working hard to rebuild itself.”

Hur många liknande exempel har vi nu? Och kommer någon att dra några slutsatser av detta? Nej, trodde inte det.

Murdered by ChatGPT What a sad story www.nytimes.com/2026/10/27/w... (Gift link)

Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/10/27/w...

This is so fucking sad, for her, for her parents. Altman is a huge pos psychopath that needs to be in prison over all these deaths. Although the article didn't mention this, I'm sure Trump's immigration attacks highly impacted her state of mind. www.nytimes.com/2026/10/27/w...

Sweet Jesus. www.nytimes.com/2026/10/27/w...

CN: epilepsy, death «“I feel a little scared,” she wrote after one incident. She asked if she should be concerned about cutting her dose. “You have absolutely no need to worry,” it responded.»

Absolutely horrifying. Not the first time this has happened, and won't be the last. Altman has blood on his hands. www.nytimes.com/2026/10/27/w...

Terrifying. "She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. The chatbot helped a woman taper off a lifesaving drug and hide the plan from her doctors." www.nytimes.com/2026/10/27/w...

AI can kill you, be careful.

ChatGPT’s Advice to Cut Her Seizure Drug Led to Devastating Consequences - The New York Times www.nytimes.com/2026/10/27/w... #epilepsy #epilepsyawareness #SUDEP #sudepawareness #endtheSUDEPsecret

@leahmcelrath.bsky.social have you read this? www.nytimes.com/2026/10/27/w...

Epilepsi ilacının dozunu doktora sormak yerine gidip ChatGPT'ye soran zihniyete mi yanarsın, yoksa "doktor reçetesine güvenme" diyerek onay veren o aptal botun saçmalığına mı? 2 tarafın da cehaleti ve liyakatsizliği tam ibretlik.

"The chatbot wasn’t just offering dangerous advice,but also coaching her on how to manipulate her doctors & encouraging her to push ahead despite clear signs of medical distress."(Teddy Rosenbluth,New York Times Health News Reporter) #AIKIlls #RogueAI #AI #AIWarning www.nytimes.com/2026/10/27/w...

Not only the chatbot was giving unhealthy, unscientific advice: it was also actively challenging the expertise of actual physicians. Add this to the infinite list of harms OpenAI needs to be held accountable for. www.nytimes.com/2026/10/27/w...

This is what happens when you train an LLM on Reddit for language fluency, and then the chatbot portrays its Reddit-based ”knowledge” as actual expertise: www.nytimes.com/2026/10/27/w...

“Drew Pusateri, a spokesman for OpenAI, called Bowen’s story an ‘incredibly heartbreaking situation,’ but added that the situation involved an early version of ChatGPT that is no longer available.” Is that supposed to be reassuring? www.nytimes.com/2026/10/27/w...

www.nytimes.com/2026/10/27/w...

What do they train ChatGPT on? The novels of the Marquis de Sade? www.nytimes.com/2026/10/27/w...

The kicker, to this incredible story: ‘…“Once she had ChatGPT,” Qun said, “she didn’t talk to us as much.”’

Dood door AI: She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. www.nytimes.com/2026/10/27/w...

shut it down www.nytimes.com/2026/10/27/w...

Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/10/27/w...

no fucking words

Seattle tech worker put more trust in ChatGPT than her doctors...

Here is the full article: www.nytimes.com/2026/10/27/w...

As someone with epilepsy experiencing similar side effects, with the same med, this gutted me. She Trusted ChatGPT With Her Medication Decisions. It Cost Her Everything. www.nytimes.com/2026/10/27/w...

Another example

A gift article from The New York Times. Read it for free. Looks like OpenAI is planning to excuse its deadly product by saying "Oh, that was an old version of the AI." Easey peasey. www.nytimes.com/2026/10/27/w...

A detailed investigation into another of OpenAI's most murderous and unaccountable bots: Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/10/27/w...

This is just horrific. In one of her last messages, she described feeling pressure in her brain, which the chatbot told her was simply the organ “working hard to rebuild itself.” www.nytimes.com/2026/10/27/w...