# Watchdog finds failures in ChatGPT's teen safety protections

_Published Saturday, October 10, 2026 at 3:18 PM EDT · AI · Latest · Tier 2 — Notable_

![A phone with the ChatGPT app in a file photograph. (Oleksii Pydsosonnii/The Epoch Times) — Primary](https://i.ntd.com/assets/uploads/2026/10/id1177994-EpochImages-7485548273.jpg-900x506.webp)

Common Sense Media's Youth AI Safety Institute said Oct. 8 that tests found failures in ChatGPT's teen enrollment and parental alerts, prompting it to recommend blocking teen access until fixes undergo independent testing.

Adult accounts were not switched to teen mode even after testers explicitly said they were 13. Four test accounts discussing suicide, self-harm or disordered eating did not trigger parental notifications. Prompts judged to suggest a crisis drew a hotline in 23 percent of cases, down from 33 percent before the program's rollout.

The institute receives funding from the OpenAI Foundation among other organizations. OpenAI did not respond to a request for comment before publication.

## Sources

- [NTD News](https://www.ntd.com/some-safety-protections-in-chatgpt-for-teens-dont-work-watchdog-finds_1177993.html)

---
Canonical: https://techandbusiness.org/newswire/ztIXME-51L4UXQIvJfO6Zl
Published: 2026-10-10T19:18:33.741Z
Story chronology: 2026-10-08T00:00:00.000Z
Retrieved: 2026-10-10T22:09:55.180Z
Publisher: Tech & Business (techandbusiness.org)