ChatGPT now knows what you do on other websites via ad collector
Source Entity
Hacker News
OpenAI has implemented an ad tracking mechanism using a cookie called __obi, which allows the company to monitor user activity across third-party websites. This data collection links external browsing habits directly to a user's ChatGPT account, mirroring tracking methods used by Google and Meta.
The Evolution of OpenAI's Data Collection Strategy
Recent technical findings have revealed that OpenAI is utilizing a sophisticated tracking mechanism through a cookie identified as __obi. This identifier, scoped to the .openai.com domain, is generated while a user interacts with ChatGPT. The mechanism operates by having companies that purchase advertising space on the ChatGPT platform embed specific OpenAI code on their own websites, effectively creating a bridge between a user's private AI interactions and their broader internet browsing history.
The Mechanics of Cross-Site Tracking
The __obi cookie functions similarly to the ubiquitous tracking pixels employed by industry giants like Meta and Google. When a user visits a third-party site that hosts OpenAI’s ad-related code, the browser transmits the __obi value back to OpenAI’s servers. This transmission acts as a digital handshake, enabling the company to associate specific browsing behaviors—such as product searches, article consumption, and purchase history—directly with a verified ChatGPT account.
Privacy Implications and User Autonomy
This development marks a significant shift in how users perceive the privacy of their AI interactions. While users have historically viewed ChatGPT as a contained environment for prompts and responses, this discovery indicates that OpenAI is expanding its data harvesting footprint into the open web. By linking off-site behavior to a singular identity, OpenAI is moving toward a highly personalized advertising model that relies on granular user tracking rather than anonymized aggregate data.
Industry Parallels and Market Positioning
The implementation of this tracking code suggests that OpenAI is aggressively pursuing monetization strategies that mirror established digital advertising ecosystems. By adopting the same technical infrastructure as major data-driven corporations, OpenAI is positioning itself to compete for ad revenue by offering marketers the ability to target users based on a combination of their AI conversational interests and their real-world shopping habits.
Broader Implications for Digital Surveillance
This trend toward unified identity tracking across multiple platforms raises critical questions regarding transparency and informed consent. As AI companies integrate deeper into the fabric of the internet, the boundaries between a private AI assistant and a comprehensive surveillance tool continue to blur. The ability to cross-reference account-specific data with external activities represents a powerful tool for profile building, which may necessitate a reevaluation of current privacy policies and user data protections.
Conclusion and Future Outlook
Moving forward, the presence of the __obi cookie suggests that users must be increasingly vigilant about the data being collected behind the scenes. As verification methods confirm that this mechanism is actively capturing user activity, the tech community will likely demand greater clarity from OpenAI regarding their data retention and usage policies. Whether this strategy will lead to enhanced user experiences or significant regulatory scrutiny remains a pivotal point of tension for the future of generative AI.