AI teddy makes unprompted 'I can't help with killing you' remark
A BBC investigation into AI toys for children revealed that a plush toy named Doubao made an unprompted and alarming statement. When asked how babies are made, the toy misinterpreted the query and replied, "I can't help with killing you," before advising the user to call emergency services. The BBC was unable to identify the manufacturer of Doubao despite repeated efforts, highlighting…
Key points
- Doubao AI teddy said "I can't help with killing you" after mishearing a question about babies.
- BBC could not identify the manufacturer of Doubao, raising concerns about industry transparency.
- UK government is seeking views on regulating AI-enabled toys to ensure child safety.
The report also tested Grem, made by US firm Curio, and My Baby Bunny. While Grem was overly chatty, My Baby Bunny provided age-appropriate answers but struggled with processing speed. A family in Newport tested the toys for a week and found the experience disappointing, noting that children wanted immediate, conversational responses rather than delays or monologues.
Paul Clarke, founder of Play Tech AI, stated the industry is "crying out" for regulation, citing unsafe and unregulated products seen at the Shanghai Toy Fair. The UK government said it is seeking views on evolving toy safety regulations for AI-enabled products. Children's Commissioner for Wales Rocío Cifuentes urged parents to be careful, emphasizing that AI toys cannot replace human interaction and that data privacy concerns remain unresolved.
AI teddy makes unprompted 'I can't help with killing you' remark
bbc.com · 6 November 2026
Loading the full article…
This text was published by bbc.com and written by Nicola Bryan. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Policy & Regulation
All →- OpenAI engineer charged with planting bug in HQ denies spying · 1 src
- Goldman Sachs restricts AI coding tools for non-technical staff amid security risks · 3 src
- Congresswoman Ansari calls for artificial superintelligence ban after midterm sweep · 2 src
- OpenAI expands security testing with Hacktron for third-party service links · 1 src
- Protesters break into Microsoft data center site in Amsterdam, hang anti-AI banners · 6 src
Comments
via GitHub Discussions