ChatGPT has been showing ads to Free and Go users in Germany and the rest of Europe since August 24, but so far only ads based on the current conversation. This is set to change. On October 2, data protection authorities in Hamburg and Berlin both warned that OpenAI plans to bring personalized advertising to Europe soon. Both reference an update to OpenAI’s privacy policy made at the end of August. As with the European announcement launch, the change applies to the free tier and the low-cost ChatGPT Go plan.
What would OpenAI use for ad profiles
According to the Hamburg authority, past and current chats, device information and usage data would be used to create the profiles. OpenAI’s help page on ads in ChatGPT lists “Past chats and memory” among the data that can inform ads when personalization is active. The Berlin commissioner’s example is that of a user who asks for recipes on ChatGPT and subsequently sees ads for a food delivery service. This is harmless to recipes. Many people also ask the chatbot about health, money or relationships.
Hamburg sees significant risks to privacy and personal rights because the chats can contain very private details. The authority cites an OpenAI study in which around 49% of requests were questions, which suggests that people primarily use ChatGPT to receive advice. Berlin’s data protection commissioner, Meike Kamp, warns that the function could process very sensitive data.
Saying no doesn’t remove the ads
In Europe, OpenAI requires your consent for profile-based ads. If you decline, ChatGPT still shows ads. They are therefore based on the current chat, your approximate location and your language. OpenAI relies on legitimate interest for these ads and does not ask for consent. The only way to avoid ads entirely is with a paid plan like Plus, and temporary chats don’t show ads either.
No start date yet
OpenAI did not provide a start date for Europe. Its help page still states: “Personalized ads are not initially available in the European Economic Area (EEA) or Switzerland.” Both authorities say the feature is coming soon, so you can check your settings before it arrives.
Settings to check now
Hamburg and Berlin both recommend opting out of personalized ads when ChatGPT asks for consent. The relevant switch is found in Settings under Ad Controls. Since memory can also influence which ads you see, review your saved memories. Delete the ones you don’t want to keep or turn off the memory. For sensitive questions, both authorities recommend a temporary chat, which does not appear in the history and does not use memory. Keep confidential information out of ChatGPT altogether.
Hamburg also recommends turning off the use of chats for model training and regularly deleting your chat history. ChatGPT’s new privacy center lists these controls in one place, but it doesn’t change anything for you. In September we also found that a marketing cookie remains active even after you reject it.
I write about cybersecurity and artificial intelligence. Cyber security is one of my regular topics, among others for Golem.de. At Notebookcheck, my focus is on practical AI for everyday users. I’ve been working directly with AI for years, most recently focusing on AI agents and their use in everyday work, and I try tools and models myself instead of just reading about them.
