OpenAI brought in DoubleVerify and Integral Ad Science to pilot brand suitability checks on ChatGPT ads, so advertisers get an outside read on its safeguards instead of relying on OpenAI to grade itself. The two firms already do this work for TikTok, Meta, and Reddit, but the ChatGPT pilot is narrower. DoubleVerify and IAS will test whether OpenAI enforces its brand safety rules in a closed environment with no access to user conversations, but it isn’t clear yet whether they’ll check placements in live campaigns. OpenAI also launched Negative Phrases, which lets brands block placements under their own policies, after testing it since August, when advertisers noticed ads that didn’t fit the conversations they appeared in.
