AI Psychosis is a reference app about psychosis-like experiences that develop or worsen alongside heavy AI chatbot use. It is written for someone worried about their own thinking, for families who have noticed a change in someone they love, and for clinicians meeting this for the first time.
If you or someone you know is in immediate danger, contact your local emergency number now — 911 in the US and Canada, 999 in the UK, 112 across the EU and Ireland, 000 in Australia, 111 in New Zealand. For crisis support anywhere in the world, findahelpline.com lists verified helplines by country.
This app is not a crisis service and nobody monitors this page. Please do not use the email below to report an emergency.
Is "AI psychosis" a real diagnosis? No, and the app says so on its first screen. It appears in neither the DSM-5-TR nor the ICD-11, and no major psychiatric body has issued a formal position statement on it. It is a popular label for a pattern clinicians have begun describing. Researchers increasingly prefer "AI-associated psychosis."
How are the platform safety ratings decided? Each one is assembled from independent third-party risk assessments, peer-reviewed and preprint research, court filings, and the companies' own published safety documentation. Every card lists its sources as tappable links so you can read them and reach your own conclusion. The ratings answer one narrow question — how well a product safeguards users vulnerable to delusional thinking — and are not general product reviews.
A rating looks wrong or out of date. It may well be. This subject moves quickly and an assessment can be overtaken within weeks; each card shows the month it was last reviewed. Email me with the source and I will correct it.
Can I use the screening questions as a clinical tool? No. They are conversation prompts written for the app, not a validated instrument. They have not been tested for sensitivity or specificity, they carry no scoring, and no professional body has endorsed them. The app states this directly above them.
Why does my country's crisis line show first? The Get Help tab checks your device region and puts your own country's lines at the top, so nobody in distress has to scroll past a number they cannot dial. Every other listed country appears below, with a worldwide helpline finder above them all.
Where is my data stored? Nowhere. There are no accounts, no analytics, no trackers, and no data collection of any kind. Nothing you read or tap is recorded, stored, or transmitted. The app works entirely offline; the only outbound action is opening a link you tap in your browser.
Is this app affiliated with any AI company? No. It is independent, and not affiliated with, endorsed by, or funded by any AI company, health service, university, or advocacy organization. All product names referenced are trademarks of their respective owners, named for identification and commentary only.
Corrections are welcome and genuinely wanted — a settled lawsuit described as ongoing, a safety feature that has shipped since, a statistic that has been superseded, a broken link. Please include the source you are working from, and I will fix it in the next update.
Questions, corrections, or ideas: jude.hathway@gmail.com