ChatGPT Ads Backlash and Enshittification

The sequence began March 22nd with something almost clinical in what it revealed. Ads appeared inside ChatGPT. Users caught it within hours. OpenAI pulled them — no blog post, no statement, no acknowledgment that anything had happened at all. Just a quiet deletion that, paradoxically, told you everything: a company that was confident in a decision doesn't erase it without a word. Then, without explanation, the ads came back. By March 24th, Cory Doctorow's word "enshittification" — the slow degradation of a platform as it shifts allegiance from users to advertisers — had already attached itself to the story and refused to let go. March 27th produced the most viscerally telling data point: one user received a ChatGPT ad notification on their phone and uninstalled the app immediately. Not "considered uninstalling." Immediately. March 28th surfaced the harder question — what happened to OpenAI's own stated position against targeted advertising? — and nobody from the company answered. By April 1st, the competitive logic had crystallised in a single sentence: why would anyone use ChatGPT with ads when every competing chatbot doesn't have them? Then, on April 20th, the supply side closed the loop. An OpenAI ad partner was publicly selling ChatGPT placements based on "prompt relevance" — meaning the content of your conversation becomes the targeting signal. You ask about flights, you get a travel ad. The chatbot that was supposed to be your honest interlocutor is now reading your queries and monetising them.

If confirmed, here is what this means. OpenAI has crossed the threshold that defines platform decay, and it has done so at exactly the wrong competitive moment. The "prompt relevance" targeting model is not a minor monetisation tweak — it is a structural conflict of interest embedded into the product's core function. A model that knows an ad partner pays for flight-related placements has a financial incentive to steer conversations toward flight-related topics. Users cannot audit this. They cannot even see it happening. The trust damage compounds fast: every answer now carries a question mark about whether the response was shaped by what you asked or by what someone paid. For OpenAI specifically, the calculus is brutal. The free tier exists to convert users to paid subscriptions. If the free tier becomes actively unpleasant — intrusive, compromised, commercially polluted — those users don't upgrade. They leave. Anthropic, Google, and a dozen open-source alternatives are one uninstall away. The users most likely to encounter these ads are also the users least committed to the platform. OpenAI is taxing its own growth engine.

Watch for OpenAI either acknowledging the targeting mechanism publicly or having a publisher or regulator force the disclosure. Any silence from the company about how "prompt relevance" actually works should be read as an answer in itself.

NewsHive monitors these sources continuously. All signal titles above link to the original reporting.

Intelligence by NewsHive. Need help navigating what this means for your business? Contact GeekyBee →