ChatGPT Ads Manager Lands Inside ChatGPT Work

ChatGPT Ads Manager now lives inside ChatGPT Work.

StartupHub data

Companies working on this

Profiles of the companies named in this story, with funding and a one-liner from our database.

OpenAI

An artificial intelligence research organization developing and promoting friendly AI for the benefit of humanity.

OpenAI

$13.0B

Artificial intelligence research and deployment company focused on developing advanced AI models like GPT-5.6 and GPT-Live, offering products such as ChatGPT and an API platform.

OpenAI

Private / $100B+ est

OpenAI is an AI research and deployment company dedicated to ensuring that artificial general intelligence benefits all of humanity.

OpenAI

$190.6B

An AI research and deployment company building safe and beneficial artificial general intelligence.

According to OpenAI Youtube, a new Ads Manager plugin lets advertisers interrogate campaign data directly in ChatGPT Work instead of pulling reports across tools. The underlying ads system stays isolated from the model's answers.

The full discussion can be found on OpenAI Youtube's YouTube channel.

The demo used Nova Electronics headphones as the example. The user intent was work, travel and home use, and the ads model delivered a Nova ad alongside the guidance.

OpenAI was explicit on the boundary. Ads do not influence ChatGPT's responses. They are clearly labeled and displayed separately.

Affected systems are ChatGPT consumer surfaces where Chat Ads appear and the advertiser stack that reads performance data. That stack previously required the API or the standalone Ads Manager. Now it includes the ChatGPT Work plugin as a third interface.

No attacker access is required to see the separation in action. The risk is not a remote one, it's a trust exploit if users or advertisers assume the model is steering recommendations toward paid placements.

How the new flow works in practice

The separation is a policy and presentation boundary, not a model behavior change.

Think of it like a pharmacy consultation with a billboard outside the window. The pharmacist's advice is the same whether the billboard is there or not, but both are visible in the same building.

In the demo, the marketer asked how three live creatives were doing: "your world, your focus, here every detail made to be yours." The plugin surfaced clickthrough rate and conversion, flagged "your world your focus" as the leader, then answered why it was ahead and what to do next.

That loop is the point. Reporting becomes a prompt, diagnosis becomes a follow-up question, and iteration on creative becomes the next turn.

Why this matters and what's still unresolved

For builders and marketing teams, this collapses the distance between measurement and making. You can query performance, diagnose the why, and draft the next creative in one thread.

It also raises the stakes on verification. A conversational summary is convenient, but teams should still reconcile against the API or Ads Manager. LLM explanations of why an ad won can be plausible and still incomplete.

OpenAI said Chat Ads are now available in 40 markets globally after launching 31 markets in Europe yesterday. Get started details are at ads.openai.com.

What is not fixed is the broader pressure this creates. Advertisers gain a faster feedback loop inside the same product where buying decisions take shape, which makes clear labeling and consistent non-influence essential to maintain trust.

The gap to watch is auditability. If the Work plugin becomes the primary way teams read ads, builders will want logs of what data was queried, what timeframe and attribution were used, and how the model's why was derived.

For now the mitigation is architectural and product level: keep ads separate from answers, label them, and offer the raw controls via API and Ads Manager for anyone who wants to check the math.