OpenAI appears to be giving advertisers more control over where their ads show in ChatGPT.
Adthena, via CMO Ashley Fletcher, says it has spotted a new “Negative Phrases” setting inside a UK advertiser’s ChatGPT Ads Manager account. The feature lets advertisers add up to 100 words or phrases that prevent an ad from appearing when a user’s message contains them.
Why This Matters:
Think keyword blocking, but for ChatGPT Ads.
Until now, advertisers have had relatively limited control over the conversations where their ads can appear. This is to be expected, of course, given how new/nascent ChatGPT Ads are. OpenAI’s contextual targeting can help steer ads toward certain topics, but negative phrases give advertisers a more explicit way to tell ChatGPT where they don’t want an ad to show.
That could be useful for both performance and brand safety. A travel advertiser, for example, could theoretically exclude conversations that signal someone has no intent to travel through a combination of negative phrases. Brands could also exclude sensitive terms or topics they don’t want associated with their advertising.
Experts React:
“That’s a real shift,” Fletcher wrote on LinkedIn. “It pulls ad placement control on ChatGPT toward the kind of precise, rule-based targeting performance marketers already rely on in Google and Bing.”
He cautioned that the rollout still appears limited and early. Here’s his full post:
Our Take:
This is another small but important sign of ChatGPT Ads growing up. (The steps the company is taking are popping up all over.)
Conversational advertising is a massive opportunity, but, of course, it creates a targeting challenge that search advertising doesn’t have to the same degree. Conversations are inherently messy, nuanced and potentially sensitive. Giving advertisers a straightforward way to say “don’t put us here” makes a lot of sense.
