Brief
OpenAI lets advertisers target ChatGPT users via email and device IDs
OpenAI's ad system can match uploaded user identifiers to ChatGPT profiles, but users can only opt out of certain ad personalization.
By Felo News Desk · Published
OpenAI has begun allowing advertisers to target individual ChatGPT users by uploading lists that contain email addresses, phone numbers or a unique Android advertising ID, Gizmodo reported on Oct. 16.
According to the report, an advertiser must provide a list of at least 25,000 entries. OpenAI then checks whether any of those identifiers belong to a ChatGPT account and, if a match is found, includes the user in the audience for that ad. The matching can use raw data, hashed versions or the device‑specific ID.
OpenAI says the advertiser never sees which specific accounts were matched, and the company claims to delete all uploaded files within 24 hours of processing. However, 404 Media notes that it is unclear whether OpenAI hashes the raw data before storage, leaving the possibility that the information could be exposed.
Users can disable personalized ads based on past chats, activity and preferences, but the setting does not block ads that rely on the current chat or the unique device identifier, according to OpenAI documentation cited by Gizmodo.
Gizmodo reached out to OpenAI for comment but did not receive a response before publication.
Key facts
- Advertisers can upload lists of at least 25,000 emails, phone numbers or Android IDs to target ChatGPT users. (gizmodo.com)
- OpenAI claims to delete uploaded identifier files within 24 hours, but it is unclear if the data is hashed first. (gizmodo.com)
- Users can opt out of ads based on past chats and preferences, but not ads based on current chat or device ID. (gizmodo.com)
Sources
- [1] gizmodo.com — originally reported as “Here's How ChatGPT Is Targeting You With Ads”
