When ChatGPT Advertising Began
OpenAI began testing advertising in ChatGPT in the United States on 9 February 2026. Its March update announced plans to expand the pilot to Canada, Australia and New Zealand. This is a developing advertising channel, so availability and controls need to be checked in the account before planning a campaign. See OpenAI's advertising announcement for the rollout history.
How Placements Differ from Search Ads
Sponsored placements are labelled and visually separate from ChatGPT's answers. Paying for an ad does not buy a recommendation in an answer. Advertisers do not receive users' private conversations or health histories.
OpenAI describes matching based on conversational context and, depending on user settings, other permitted signals. This does not give a clinic access to condition-based audience lists or a way to select individual patients.
Healthcare Placement Restrictions
OpenAI's published test rules restrict ads near sensitive or regulated topics, including health and mental health. A user asking about symptoms or treatment is therefore not an audience placement we can promise to reach. We check the current rules for the business and proposed creative before recommending any spend.
Assessing a Test
We review the service, intended audience, existing marketing activity and advertiser access. If the proposed campaign depends on restricted health conversations, we do not build a plan around that assumption. Where an eligible opportunity exists, we agree a limited test with clear reporting and review points.
Before launch, we agree a budget, the enquiries your team wants to receive and a date to review the test.
Advertising Content and Compliance
Australian healthcare advertising obligations apply regardless of the channel. For regulated health services, claims need evidence, practitioner credentials must be accurate, and advertising must not use prohibited clinical testimonials. Prescription medicines cannot be promoted to the public.
We review both the proposed ad and its landing page. Platform approval is not a substitute for meeting Ahpra, TGA or other applicable requirements. The assessment depends on the service being advertised and the content people will see after clicking.
Measurement and Privacy
We agree which outcomes can be measured before a test starts. Reporting can compare spend, clicks and permitted website enquiry signals with enquiries recorded by the business. Consent choices, platform restrictions and the booking system can limit attribution.
Tracking must not send health details, patient records or sensitive form responses to an advertising platform. We do not promise complete booking attribution or access to the private conversation that preceded a click.
Reviewing the Outcome
A test should answer a practical question: does the eligible campaign produce useful enquiries at a cost the business can sustain? We review enquiry quality alongside volume and cost. Your team's feedback helps distinguish a general question from an enquiry that leads to an appointment.
Related search services
For location and map visibility, see our Google Business Profile and local SEO service. For organic mentions and citations in AI answers, see AI search visibility for healthcare.
