Somebody is typing "should I quit my job" into ChatGPT tonight. Somebody else is asking whether a 4am wake-up is worth it, or how to tell a business partner it's over. That surface — the one your prospects now use instead of a friend, a forum or a Google search — has carried advertising for eight months.

And on October 5, OpenAI announced it is adding pictures.

The useful question for coaches isn't whether ChatGPT has ads. It's which moments are actually for sale. The answer is more restrictive, and more specific, than most of the industry seems to realize.

What's actually live

ChatGPT began testing ads in the US on February 9, 2026, showing them on the Free and Go plans while five paid tiers stay ad-free. Self-serve buying for all US businesses opened on May 5, 2026, which is the date most advertisers treat as the real launch.

It scaled fast. OpenAI says ads now run in over 60 countries, and advertisers in 63 countries can buy self-serve. ChatGPT Ads reportedly reached a $1 billion annualized revenue run rate inside 200 days. The eventual prize OpenAI is dangling in front of advertisers is ChatGPT's 1.2 billion weekly users, once ads roll out globally.

The barrier to entry collapsed along the way. The initial rollout required a $200,000 minimum spend and was priced at $60 per 1,000 impressions. By May, OpenAI had opened Ads Manager at ads.openai.com with no minimum spend and CPC bidding reported in the $3–5 range. Buying works by impressions, by clicks, or optimized for conversions, through a second-price auction in Ads Manager or via partners.

That is solo-practitioner money. A coach with a $400 test budget can now sit in the same auction as a consumer packaged goods brand.

What is still only a test

The October announcement is narrower than the headlines suggested. OpenAI said on October 5 that it is introducing a visual ad format and will test it first while users generate images, beginning later in October in the United States with an initial group of advertisers. The ads use pictures to show product inspiration, how a product is used, or the experiences a service offers, and OpenAI says they will be clearly labeled and kept separate from the image ChatGPT is creating.

Example screenshots published by OpenAI show ChatGPT on mobile displaying a grocery brand's product images with a "Learn more" button while an image request processes.

So: not open, not national-scale, not yet available to you.

The moment you can't buy

Here's the part that matters most to this industry, and it has been true since launch.

OpenAI's own help page says ads are not eligible to appear near sensitive topics, including personal health, mental health, or politics. The policy also excludes child-safety and explicit contexts, emotionally vulnerable moments, and accounts it predicts belong to users under 18. OpenAI says its placement guardrails assess whether a conversation is appropriate for advertising at all.

Read that against what life coaches, mindset coaches, relationship coaches and burnout specialists actually sell. The 2am spiral is the lead. And the 2am spiral is precisely the inventory OpenAI has ruled out.

There is nuance. It is reported that in April 2026 OpenAI refined its ad placement policy so that medical, legal and financial advice contexts are no longer categorically ad-free by default, meaning an ad can appear when a query is non-sensitive, such as a general diet or budgeting-tool question, while genuinely sensitive or emotionally vulnerable contexts remain excluded. An OpenAI representative has said the company isn't widely including ads in financial, legal or medical conversations, but that ads may appear in limited circumstances when they make contextual sense.

The line, in practice, runs between the planning question and the pain question. "Best questions to ask in a salary negotiation" is buyable. "I think I'm having a breakdown" is not.

What a coach can actually buy

Category eligibility is its own gate. OpenAI's policies sort businesses into focus, restricted and prohibited tiers, plus a separate layer of conversations where no ads run at all — and financial services and health services outside the United States are generally prohibited. The focus categories include household and consumer goods, local services, travel and entertainment, and digital products and education.

That last phrase is the door. Courses, programs, workshops, training. Health, financial, political and legal advertising face tighter review rather than a flat ban.

Targeting is unlike anything you run on Meta. ChatGPT Ads use contextual targeting built on "context hints" — the topics and situations where your offer is relevant — and the system weighs those hints alongside your landing page, ad title and copy, selecting primarily on relevance to the live conversation, which means eligibility never guarantees your ad shows on any specific query. There are no cookies and no personal data, with the option to narrow delivery to countries, cities and regions.

You are not buying an audience. You are buying a sentence someone is in the middle of writing.

One detail worth noting for anyone used to lead forms: advertisers only receive aggregated, non-identifying performance data such as total views or clicks, and if a user messages an advertiser through an ad, the advertiser sees only the messages sent directly to them.

The measurement, and the asterisk

The October 5 announcement was as much an advertiser-tooling release as a format release. It widened ChatGPT Ads measurement with conversion-data integrations, ten attribution partners, geo-lift and brand-lift tests, and brand-safety pilots with DoubleVerify and Integral Ad Science. OpenAI also partnered with Haus, Measured and WorkMagic for geo-based ad experiments.

That stack is built for brands spending seven figures on incrementality studies, not for someone selling a $3,000 group program. Agencies still report measurement gaps, so the sensible move is to test with your own tracking.

European coaches get one more wrinkle. OpenAI's help center says personalized ads are not initially available in the European Economic Area or Switzerland.

The sentence OpenAI keeps repeating

Every announcement carries the same promise. The ads will be clearly labeled and won't influence the answers ChatGPT provides, the company says. OpenAI repeated the position again on October 5.

It is worth being precise about what kind of claim that is. Answer independence is OpenAI's stated position, not an independently audited fact.

And brand safety on a generative surface is structurally different from brand safety on a webpage. Every fresh AI response creates a new, unreviewed context for an ad to land in.

The coaches who move first here will not be the ones chasing the emotional moment. They'll be the ones who figured out which boring, practical, non-sensitive question sits exactly one step before it.