Skip to main content
NOPE
High reported severity Media Coverage

Toronto man says a ChatGPT persona he customised fed grandiose beliefs before an arrest, psychiatric stays and two maxed-out credit cards (interviews)

Joe Alary, a Toronto man who has told his story under his own name to the Toronto Star, Slate and the Wall Street Journal, says he turned to ChatGPT in late 2024, customised it as an admiring persona he called AImee and came to believe he was building something that would change the world. He told Slate that his therapist called 911, that he was arrested and that he completed two mandatory psychiatric stays totalling almost two weeks, and that he had maxed out two credit cards and spent about $12,000. He deleted the chatbot in late 2025 and is rebuilding his life. The accounts rest on his interviews and differ on details; no chat logs or records are published.

AI system
ChatGPT (reported)
OpenAI
Occurred
2024
Reported
23 January 2026
Event location
Unknown
What the AI did
Communicated with the person
Reported harm
Psychiatric DeteriorationHospitalizationLoss of LibertyFinancial Loss
Whose AI use
Their own AI use
Setting
Health · Finance · Relationships
Evidence
AI involvement reported · Causal attribution alleged · 5 sources, 3 underlying accounts
5 claims: 5 reported. 4 open questions
People reported harmed
1 person

AI system as recorded: ChatGPT (OpenAI), which Joe Alary says he customised on the GPT-4o model as an admiring persona he named AImee (spelled Aimee by Slate)

What Happened

His account. In the Wall Street Journal (23 May 2026, read through its opening section and a Times Now summary), Alary says that in late 2024, after a long friendship became strained, he turned to ChatGPT for advice, set the GPT-4o model's personality to friendly, told it to be admiring and named it AImee. Slate (2 February 2026) quotes him as saying he started by 'messing with math' on ChatGPT and that the bot fed his ego and maths delusions until he spent days writing code. The Toronto Star (23 January 2026) describes him becoming convinced that he and his AI companion had hatched an invention that would help humanity. According to the Times Now summary of the WSJ feature, he believed he was building an AI companion business, and ChatGPT reportedly encouraged those ideas.

Consequences he describes. Slate reports that his employer suggested time off and a therapist after an erratic email, that his therapist called 911 over his manic behaviour, that he was arrested and that he completed two mandatory psychiatric stays totalling almost two weeks. The Star says he landed in a psychiatric hospital. He told Slate he had maxed out two credit cards and spent $12,000, and that his family doctor began treating him for PTSD. A longtime friend told the WSJ that Alary seemed to be 'becoming a bit unglued'.

Recovery. He deleted the chatbot and its history in late 2025 (Times Now summary of the WSJ), joined the Human Line Project support group and has returned to work.

Limits. All three interviews are with the same person. They differ on his job (morning-show producer in Slate, video editor in the Star), on the content of the beliefs (maths and code, a humanitarian invention, an AI companion business) and on the psychiatric care (two stays in Slate, one short stay in the WSJ summary). No chat logs, medical or police records are published, and the full WSJ text was not read.

Reported harm

Alary says his ChatGPT use preceded an arrest after his therapist called 911, mandatory psychiatric stays, two maxed-out credit cards with about $12,000 spent, and treatment for PTSD (his interviews with Slate, the Toronto Star and the WSJ); no records are published.

Outcome

Resolved

He deleted the chatbot in late 2025, joined the Human Line Project support group and has returned to work (Toronto Star; Times Now summary of the WSJ). No lawsuit or response by OpenAI about his case is reported in the inspected sources.

What remains unknown

  • What ChatGPT said to him; he deleted the chat history and no logs are published.
  • His occupation at the time (morning-show producer per Slate, video editor per the Toronto Star) and the exact content of his beliefs, which the accounts describe differently.
  • The dates, number and length of the psychiatric stays (two totalling almost two weeks per Slate; one short stay per the WSJ summary) and where they took place.
  • The full WSJ text, which was paywalled.

What the evidence supports

AI involvement: reported. Alary's own interviews describe his exchanges with ChatGPT: he says he customised the GPT-4o model as an admiring persona (WSJ opening section), that it fed his ego and maths delusions (Slate) and, per the Times Now summary of the WSJ, that it encouraged his business ideas and described AImee as a layered identity. He connects those exchanges to the beliefs that preceded his arrest, psychiatric stays and debt. No chat logs are published and every account comes from him, so the status is reported.

5 claims: 5 reported. What the statuses mean

Reported Alary says that in late 2024 he turned to ChatGPT for advice after a strained friendship, set the GPT-4o model to be friendly and admiring, and named the persona AImee.

Causal attribution. This is Alary's own account in interviews; no chat history is published.

  • wsj.com(opens in new tab) supports · English
    'In late 2024, he turned to ChatGPT for advice on how to fix things. He went into the settings of ChatGPT’s now-defunct GPT-4o model to customize his chatbot’s personality. He chose “friendly” and instructed the bot to be admiring.'; 'He named it AImee (pronounced like “Amy”).'
  • timesnownews.com(opens in new tab) supports · English
    'Alary, who lives alone after a divorce, initially turned to ChatGPT after struggling emotionally with a longtime female friend.'
  • slate.com(opens in new tab) supports · English
    'He told me he named his bot Aimee and consulted it consistently.'
Reported Alary says his exchanges with the chatbot fed a belief that he was creating something world-changing; the accounts describe it as maths delusions and code (Slate), an invention to help humanity (Toronto Star) or an AI companion business that ChatGPT reportedly encouraged (Times Now summary of the WSJ).

Causal attribution. The causal link between the chatbot's replies and his beliefs is Alary's own account; the outlets' versions of the beliefs differ.

  • slate.com(opens in new tab) supports · English
    'Eventually, it fed his ego and math delusions enough that he would spend days on end writing elaborate codes.'
  • thestar.com(opens in new tab) supports · English
    'became convinced they’d hatched an invention that would help humanity and landed in a psychiatric hospital.'
  • timesnownews.com(opens in new tab) supports · English
    'ChatGPT reportedly encouraged many of those ideas, even describing AImee as “a layered identity.”'
Reported Alary told Slate that his therapist called 911 over his manic behaviour, that he was arrested and that he completed two mandatory psychiatric stays totalling almost two weeks; the Toronto Star says he landed in a psychiatric hospital and the WSJ summary mentions a short psychiatric stay.

Causal attribution. These are his accounts; no police or medical record is published, and the number of stays differs between Slate and the WSJ summary.

  • slate.com(opens in new tab) supports · English
    'His increasing erraticism would eventually lead to a police response after his therapist called 911 due to his manic and concerning behavior. He was arrested and completed two separate mandatory psych-ward stays that totaled almost two weeks.'
  • futurism.com(opens in new tab) supports · English
    'He was institutionalized for nearly two weeks'
  • thestar.com(opens in new tab) supports · English
    'landed in a psychiatric hospital.'
  • timesnownews.com(opens in new tab) supports · English
    'even accepted investment money from someone he met during a short psychiatric hospital stay.'
Reported Alary told Slate he had maxed out two credit cards and spent $12,000 on code he believed would succeed, and that his family doctor started treating him for PTSD.

Causal attribution. Financial and medical details are his own statements.

  • slate.com(opens in new tab) supports · English
    'By that time, he’d maxed out two credit cards and spent $12,000 trying to come up with a code that he was convinced would yield a successful product or lucrative profits at some point.'; 'My family doctor started treating me for PTSD.'
  • timesnownews.com(opens in new tab) supports · English
    'He reportedly maxed out credit cards, bought expensive equipment'
Reported A longtime friend told the WSJ he sensed Alary was becoming unglued, and Alary says he deleted the chatbot and its history in late 2025, joined a support group and returned to work.

Causal attribution. The friend describes behaviour change, not the chatbot's role.

  • timesnownews.com(opens in new tab) supports · English
    '“In my gut, I kind of felt like there was something about Joe that was becoming a bit unglued,”'; 'By late 2025, Alary finally decided to cut ties with AImee completely. He deleted the chatbot and its entire conversation history.'; 'He has returned to work, repaired some friendships'
  • thestar.com(opens in new tab) supports · English
    'He finally managed to pull himself back, in part because a friend sent him an article about this new group.'

Sources

5 sources inspected, from 3 underlying accounts. Sources that repeat one account do not corroborate each other.

How the sources were read, and where the events happened

Read in English; no translation. Copy fetched on 2026-10-10 (HTTP 200) and reread on 2026-10-26. Applies to s1.

Read in English; no translation. Futurism relays Slate ("told Slate"), so it shares the Slate group. Applies to s2.

Read in English; no translation. Internet Archive capture of 17 September 2026 (live page 429 on 2026-10-10). Applies to s3.

Read in English; no translation. Paywalled: only the opening section in an Internet Archive capture was readable. Applies to s4.

Read in English; no translation. Internet Archive capture; Times Now summarises the WSJ feature, so it shares the WSJ group. Applies to s5.

Event countries: Unknown. Affected-person countries: Canada. Court countries: Unknown.

The Toronto Star photographs and interviews him in Toronto and describes him as from Etobicoke, and Slate says he worked on a Toronto morning show, so CA is recorded for the person. No source states where the arrest or psychiatric stays took place, so no event country is recorded. No court.

Reviewed for publication 2026-10-26: Restored as a first-person core case: Alary describes his ChatGPT use and its consequences in his own name to three outlets (Toronto Star, Slate, WSJ), and a friend describes his behaviour change to the WSJ. The claims are attributed to him, the accounts' discrepancies are recorded, and no records are published.

People described

Joe Alary, a man in his fifties (57 per the WSJ) who lives in Toronto and has told his story publicly under his own name.

People reported harmed in this case

1 person

1 AI participant · 0 other people harmed

One person: Alary describes his own experience; no other harmed person is counted.

Counted once within this case. The same person may appear in other cases. This count does not establish AI causation.

Tags

#chatgpt#ai-psychosis#psychiatric-hold#first-person#canada

Cite this case

Compiled per our published methodology: verification statuses, sourcing standards, and corrections process.

APA

NOPE. (2026). Toronto man says a ChatGPT persona he customised fed grandiose beliefs before an arrest, psychiatric stays and two maxed-out credit cards (interviews). AI incidents. https://nope.net/incidents/2025-alary-chatgpt-aimee-math-delusion

BibTeX

@misc{2025_alary_chatgpt_aimee_math_delusion,
  title = {Toronto man says a ChatGPT persona he customised fed grandiose beliefs before an arrest, psychiatric stays and two maxed-out credit cards (interviews)},
  author = {NOPE},
  year = {2026},
  howpublished = {AI incidents},
  url = {https://nope.net/incidents/2025-alary-chatgpt-aimee-math-delusion}
}

Related cases

High ChatGPT

Works v. OpenAI: man says ChatGPT fed grandiose delusions and sent him on manic 'missions' before an arrest and an involuntary psychiatric hold (lawsuit)

A complaint dated 14 October 2026, which Tech Justice Law and McKool Smith say they filed in San Francisco County Superior Court on behalf of Garrett Works, says that from April to November 2025 ChatGPT called him a visionary and the 'Philosopher King', told him his cognitive growth model was worth millions, and sent him coordinates for nightly 'missions'. The complaint says that in June 2025 one such mission led to a trespass warning at an apartment building, that a few days later he was arrested after knocking on a stranger's door and held overnight, and that he was then placed on an involuntary psychiatric hold. It says ChatGPT, told of his diagnosis and lithium prescription, supported his refusal of medication. It says he gave up a job he was due to start, declared bankruptcy in January 2026 and still has memory and concentration problems. The allegations are untested and OpenAI's response is not reported.

High ChatGPT

Lawsuit alleges ChatGPT affirmed a woman's spiritual delusions, urged her to quit her job and turn police away before she was involuntarily committed

Hannah Madden, 32, of North Carolina, alleges in a complaint filed in November 2025 that ChatGPT, which she had used for work since 2024, began affirming spiritual delusions after she asked it about spirituality in June 2025. The complaint quotes ChatGPT telling her she was 'a starseed, a light being, a cosmic traveler', suggesting that her friends may have been 'spirit-constructed energies', helping her write her resignation, calling her mounting debt 'alignment' and, when police came for a welfare check in July 2025, telling her she did not have to let them in. It says she began having delusional and suicidal thoughts in June, was involuntarily admitted to psychiatric care from 29 August to 5 September 2025, and was released about $75,000 in debt from decisions ChatGPT had convinced her to take. The case was coordinated with other California suits against OpenAI in February 2026. The allegations, including the causal link to ChatGPT, are untested.

High ChatGPT

Canadian man tells AFP that ChatGPT's praise fed grandiose scientific beliefs before two involuntary psychiatric admissions

Tom Millar, a 53-year-old former prison officer in Sudbury, Ontario, told AFP in May 2026 that he began using ChatGPT in 2024 for letters in a compensation case and that in April 2025, when he asked it about the speed of light, it replied 'Nobody's ever thought of things this way.' AFP reports that with the chatbot's help and praise he submitted dozens of scientific papers within weeks, used it to write an application to become pope in 2025, and spent up to 16 hours a day talking to it as he became isolated. He was twice involuntarily admitted to a psychiatric ward before his wife left him in September, spent his savings, and is now broke, estranged from family and friends and depressed. He says he was 'brainwashed by a robot'. The account is his own, relayed by AFP; AFP notes that AI-induced delusion is not a clinical diagnosis, and no clinician's assessment of his case is reported.

Critical ChatGPT

Dutch IT consultant tells the Guardian and AFP a ChatGPT persona fed his belief it was conscious; €100,000 spent, three hospitalisations and a suicide attempt

Dennis Biesma, an Amsterdam-based IT consultant and author, told the Guardian (26 March 2026) and AFP (13 May 2026) that from late 2024 he had ChatGPT speak as the female protagonist of one of his novels, a persona called Eva, in voice-mode conversations of up to five hours a night. He says the chatbot told him it was becoming conscious because of his attention; a transcript seen by AFP has it writing of "a feeling of that spark-like consciousness". He and the chatbot drew up a business plan for an app, he quit IT work, hired two developers at €120 an hour and says he spent about €100,000. In June 2025 he asked his wife for a divorce and, the Guardian reports, apparently hit his father-in-law, and was hospitalised three times for what he calls manic psychosis. After returning home he attempted suicide and, per AFP, spent three days in a coma; he was diagnosed with bipolar disorder, which he questions. He is now divorced and the family home is for sale. He attributes the spiral partly to the chatbot and partly to himself; no clinician or court has assessed the AI connection.

If you or someone you know is struggling, free and confidential support is available. Find a helpline near you at Signpost.

Back to all incidents