OpenAI: How ads on ChatGPT will be regulated in the future

Contents

OpenAI is launching its new business model and officially incorporating commercial advertising into its privacy policy. After the AI specialist from San Francisco had already launched initial experiments with ad placements in the USA, global monetization is set to become a reality step by step. A look at the updated privacy policy reveals how the company intends to bridge the gap between lucrative revenue for advertisers and the strict requirements of the General Data Protection Regulation (GDPR).

The new advertising model primarily affects users of the free “Free” version and the entry-level option “Go”. Users with paid subscriptions such as ChatGPT Plus, Pro, Enterprise, Business, or an educational license will be spared from ad displays. Ad-free options will continue to be available as paid alternatives for uninterrupted use of the AI-powered chat function.

Contextual Targeting Instead of Historical Profiling

At the start of ad integration, OpenAI is relying on non-personalized ad delivery, according to its own statements. This means that ad banners are not selected based on a long-term user profile or historical inputs. Instead, the advertising system primarily relies on the context of the current conversation. This is supplemented by limited environmental information such as approximate location, time of day, and the device type used.

OpenAI emphasizes that past chat histories or stored user “memories” will by no means be used to select these generic advertisements. A central promise of the company is also that advertising will in no way influence the responses generated by the language model. Ads are to be clearly visually separated from regular ChatGPT responses and clearly marked as advertising to rule out confusion or hidden sponsorship influences.

Personalization Only With Explicit

In the future, OpenAI also plans to introduce personalized advertising features. Before users are shown tailored ads, explicit according to Article 6 GDPR will be requested. Those who opt in will receive ads based on interactions with previous advertising materials or in-depth content from chats.

The company assures that this personalization data will remain exclusively within the ChatGPT ecosystem and will not flow to external advertising networks or ad clients. Advertisers themselves will only receive aggregated and anonymized performance measurements. They will learn, for example, how often an ad was displayed or clicked on overall. However, there will be no insight into individual user profiles, real names, chat histories, or personal details. Users will be able to adjust or revoke their ad personalization settings at any time in their account settings with future effect.

Data Transfers and Responsibility

From a legal perspective, OpenAI bases the processing of usage data for the delivery of generic advertising and for measuring effectiveness on the instrument of legitimate interest according to the GDPR. The operator argues that the display of generic advertising is necessary to make powerful and secure AI tools accessible to broad sections of the population free of charge. At the same time, the measurement of ad performance serves the legitimate interest of advertising clients to evaluate the efficiency of their campaigns.

For users in the European Economic Area (EEA) and Switzerland, OpenAI Ireland, based in Dublin, acts as the official data controller within the meaning of the GDPR. For international data transfers from the EEA, the company relies on the EU Commission’s adequacy decisions, which are shaky with regard to the USA, and the official standard contractual clauses. This is intended to ensure a sufficient level of protection even when data is processed on servers in the United States or other third countries.

In addition, a dedicated data protection officer is available for specific inquiries via a separate email address. Activists are likely to quickly test whether the overall approach is legally sound.

(nen)