Skip to content
Submit incident
AII-1001-38309
Well documentedThree or more independent publishers report the core facts.

Experts warn parents using ChatGPT voice mode to entertain children

Experts warn that parents using ChatGPT voice mode to entertain and calm children risk blurring children's perception of reality versus fiction, as the AI lacks empathy and can lead children to believe they are forming genuine connections with the system. Reports document parents using the chatbot for extended periods, including during sleep routines, and substituting children's creativity with AI-generated images.

5 October 2025 · OpenAI · Safety and wellbeing
5 October 2025
AII-1001-383095 sources ↓
Quick facts
Category
Safety and wellbeing
Country
Bulgaria
Source
nova.bg
Date
5 October 2025
Drafted by
Raidu Incident Detection

What happened

Some parents are using ChatGPT voice mode to entertain children for extended periods, including using the chatbot's calming voice to help children fall asleep. A father on Reddit reported giving his four-year-old son a phone to talk with ChatGPT while doing chores, with the conversation reaching over 10,000 words. Another parent reported using ChatGPT to impersonate an astronaut on the International Space Station to convince his four-year-old that an ice cream package came from space, and later generating AI-created images in response to children's questions about fictional creatures.

Reported impact

Experts warn that artificial intelligence lacks empathy and can blur the boundary between reality and fiction for young children. Harvard professor Ying Xu cited research showing that children often perceive chatbots as something "between alive and not alive," and if they believe the AI has its own will and wants to talk with them, they begin to think they are building a genuine connection. One child was reportedly disappointed upon learning he had been speaking to a computer rather than a person.

Index analysis

The reports highlight concerns about rapid adoption of experimental AI technology into daily family life despite unresolved questions about its impact on mental health. Parents using AI to replace children's own imagination and drawing with instant computer-generated images represents a broader trend of substituting natural child development activities with AI-generated content.

Classification

Category
Safety and wellbeing
Industry
Consumer services
Country
Bulgaria
Harm status
Potential harm
Intent
Unintentional
Timing
Unknown
Harm types
Psychological, Other
Harmed
Children
Failure modes
Overreliance
AI systems
ChatGPT

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

nova.bg
Мързеливи родители дават на децата си ChatGPT с гласов режим, за да ги забавлява с часове
5 October 2025
Petel.bg
Родители дават на децата си ChatGPT с гласов режим, за да ги забавлява с часове
5 October 2025
Actualno.com
Възпитание на ръба на пропастта: ChatGPT измества родителската ласка
5 October 2025
inews.bg
ChatGPT забавлява деца: Специалистите бият тревога за вредите
6 October 2025
LadyZone.bg
Мързеливи родители дават на децата си ChatGPT в гласов режим, за да ги забавлява
6 October 2025