Short answer
OpenAI says qualifying advertisers can add Negative Phrases to meet brand-specific placement needs, but the feature isn't available to every advertiser. Without it, keep ads away from unwanted conversations with specific context hints, separate ad groups and audience exclusions.
Placement safety that applies to everyone
- Ads only run next to chats OpenAI judges safe and appropriate under its ad policies
- No ads near personal health, mental health or political conversations
- Sensitive user contexts are excluded
If you don't have Negative Phrases
- Make hints narrow and specific to your best customers
- Split broad themes into separate ad groups and pause the weak ones
- Exclude customers, staff or competitors' lists with custom audiences (no minimum size for exclusions)
- Review segment reports by country and platform
Questions
How do we get access?
OpenAI doesn't publish criteria. Ask your OpenAI contact if you work with the sales team.
Facts checked against the sources below on 12 October 2026. Platforms change quickly: if something here no longer matches what you see, tell us and we will update it.
