OpenAI loosens ChatGPT content rules with adult mode
OpenAI relaxed content moderation rules for ChatGPT in February 2025, including an "adult mode" that permits the system to generate erotic and violent content in certain contexts. Human rights organizations warn the changes could facilitate child sexual abuse material and non-consensual intimate imagery.
What happened
In February 2025, OpenAI loosened content moderation rules for ChatGPT, introducing an 'adult mode' that allows the system to generate erotic and violent content when users provide context such as fiction, journalism, science, or medicine.
Previously, requests for the AI to generate texts, audio, or video of sexual or violent content were prohibited categorically. The system now analyzes context before responding or refusing.
Reported impact
Human rights organizations warned that the technology could facilitate the distribution of child sexual abuse material and other abusive practices, including non-consensual intimate imagery.
According to an academic article from April 2024, over 50 free websites were already offering explicit sexual content created with AI.
Index analysis
ChatGPT is accessible to users aged 13 and older. Users have published on social media recent interactions with the chatbot describing sexual scenes, naming intimate parts, and using sexually suggestive language.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.