There is a certain art to setting a table. You do not put the ketchup bottle out for guests. You decant. You garnish. You make the whole thing look effortless, even when it very much is not.
OpenAI, it seems, has decided to put the ketchup bottle out.
This week the company confirmed it is expanding its ChatGPT advertising test to five new countries — the United Kingdom, Japan, Brazil, Mexico and South Korea — after quietly starting in the United States back in February. The ads appear for logged-in adult users on the Free and Go tiers. Paying guests on Plus, Pro, Business, Enterprise and Education are, for now, spared the clutter.
A little something extra you did not ask for
Let me be clear about what is happening here. You open ChatGPT to ask a perfectly reasonable question — how to braise a short rib, how to word a difficult email — and now, somewhere in that exchange, a brand would like a word. It is the digital equivalent of a caterer handing your dinner guests a coupon between courses.
It is the digital equivalent of a caterer handing your dinner guests a coupon between courses.
OpenAI has, to its credit, laid the good linen over it. The company says advertisements will not influence ChatGPT's answers, that your conversations stay private from advertisers, and that you can dismiss an ad, tell it why the ad was shown, delete the ad data with a single tap, and review an "Ads History" of every brand you have seen. It is a very tidy apology for a thing nobody requested.
And one does understand the impulse. ChatGPT crossed a billion monthly users this summer, and a billion people is an extraordinary amount of house to keep. The compute bills are, by all accounts, obscene. When the household budget runs this hot, someone eventually suggests renting out a room.
The whole neighborhood is redecorating
OpenAI is not alone at this. Google is stitching Gemini — which just passed a billion users of its own — deeper into Search and its ad machine. The entire consumer-AI street is quietly wondering how to monetize the guest who never intended to pay. This is the year the free chatbot learned to sell.
Here is my quiet, savage little observation: the moment your assistant is paid by someone other than you, you are no longer entirely sure whose taste it is serving. Even when the answer is honest, the room now has a sponsor in it. That is not a flaw in ChatGPT. It is the floor plan of ad-supported anything.
Ask the whole table, not the one with a sponsor
The cleanest fix for a sponsored answer is a second opinion — and a third, and a thirtieth. That is precisely what Gangsta AI is for: put your question to more than 30 leading models at once — ChatGPT, Claude, Gemini, Grok — and get a single, cross-checked verdict, with no advertiser seated at the table steering the reply.
When one voice comes with commercials, ask for the consensus of the whole room. It is, as I like to say, a good thing.
Sources
- OpenAI — Testing ads in ChatGPT (official)
- gHacks — OpenAI expands ChatGPT ads test to UK, Mexico, Brazil, Japan and South Korea
- MM+M — OpenAI kicks off test phase for ChatGPT ads
- Hero photo: Office of the Prime Minister of Japan, via Wikimedia Commons (CC BY 4.0)
Originally published on Gangsta AI News. Gangsta AI lets you compare 30+ AI models side by side on a single prompt — free, no login. Try the AI model comparison tool.
Top comments (0)