Platform News

OpenAI to Test Image Ads in ChatGPT, U.S. Rollout This Month

OpenAI moves to make ChatGPT ads more measurable — and more visual
OpenAI moves to make ChatGPT ads more measurable — and more visualJacekBen / pixabay

OpenAI will test image ads inside ChatGPT in the U.S. later this month, seven months after the company began selling ads in its chatbot in February. The release also adds a measurement upgrade.

By Nathan Brooks

2 min read

At a glance

  • OpenAI started selling ads in ChatGPT in February, roughly seven months before the new image-ad test.
  • Image-ad testing inside ChatGPT begins later this month in the United States.
  • The current ChatGPT ad unit includes six components: advertiser name, logo, headline, description, landing page URL and a single image.
  • Advertisers have raised three complaints: measurement, creative formats, and inventory.
  • This release addresses measurement and formats; inventory depth is not part of the update.

OpenAI will begin testing image-based ads inside ChatGPT in the United States later this month, roughly seven months after the company first started selling advertising inside its chatbot in February.

The format change is one of two product moves OpenAI is rolling out to address advertiser feedback. The second — a measurement upgrade — arrives in the same release window, though OpenAI has not yet published technical specifics for buyers.

What advertisers have been asking for

Since OpenAI began selling ChatGPT ads in February, media buyers have surfaced three recurring complaints, according to Digiday's reporting: inadequate measurement, a narrow set of creative formats, and insufficient inventory to absorb the spend advertisers want to commit.

The latest release targets the first two of those issues directly.

The format change

For the first seven months, advertisers had access to a single ad template. That template included an advertiser name, logo, headline, description, landing page URL and one image asset. The image-ad test beginning later this month in the U.S. will expand the visual real estate available inside ChatGPT conversations.

The format expansion reflects a simple product reality: a thin initial unit limits what creative teams can do, and richer formats tend to follow once a platform has enough demand to justify the engineering work.

What this means for practitioners

Brands already running ChatGPT campaigns should expect new creative specifications to land in their ad accounts within weeks. Creative teams need to prepare for new aspect ratios, asset combinations and likely new approval workflows once OpenAI publishes the formal guidance.

Agencies still evaluating entry will be watching early performance signals from the image format before committing budget. A format upgrade alone does not close measurement gaps, but it removes one barrier to pilot spending.

The inventory question will determine how much the upgrade matters at scale. Even a stronger ad unit cannot absorb demand if supply stays capped.

What's still unaddressed

The release confirms measurement is on the product roadmap, but the source provides limited detail on what changes buyers will see in their dashboards. The third buyer concern — inventory depth — does not appear in this update.

OpenAI's ad business is roughly seven months old. Each release so far has tracked closely with what buyers were willing to tolerate, and the next several quarters will likely determine whether the platform can absorb incremental demand or whether advertisers shift spend back to more mature channels.

Source: Digiday

Nathan Brooks

Show full bio

Staff writer covering companies, deals and strategy in Social platform news and policy at Reach Brief.

More from Nathan Brooks →

Related articles

« Previous articleNext article »