ChatGPT Ads Brand Safety and Restricted Categories

Most mainstream businesses can run ChatGPT Ads, but several categories cannot. The ChatGPT Ads restricted categories include adult content, alcohol, tobacco, and gambling, while health, financial, political, and legal advertising face tighter review. OpenAI also keeps ads out of sensitive conversations like personal health, mental health, and politics. Review happens at the advertiser, ad, and placement level, so eligibility depends on your business, your creative, and the live conversation context.

What Categories Can Advertise on ChatGPT?

Most mainstream commercial categories are eligible to advertise on ChatGPT.

If you sell products or services that would run without issue on Google Search or Meta, you are very likely eligible here too. That covers the large majority of small and mid-size advertisers: retail and ecommerce, software and apps, professional services, home services, travel, hospitality, education providers, consumer brands, and business-to-business offers.

Eligibility is not the same as guaranteed delivery. ChatGPT Ads use contextual targeting, which OpenAI describes through "context hints," the topics and conversations where your offer is relevant. The system weighs those hints alongside your landing page, ad title, and ad copy, then selects ads primarily on relevance to the live conversation. Being in an allowed category means you can compete for placements. It does not promise your ad will show on any specific query.

The practical takeaway: if your business is mainstream and your offer is clearly commercial, assume you are eligible and focus your energy on relevance. The categories that need a closer look are the restricted and excluded ones covered next.

Which Categories Are Restricted or Excluded?

Some categories are excluded outright, and others are permitted only under tighter review.

The excluded categories are the ones you would expect from any major ad platform: adult content, alcohol, tobacco, and gambling. These face the heaviest restriction and are generally kept off the platform rather than approved with conditions.

A second group is restricted, meaning eligible in principle but subject to tighter review. This includes health, financial, political, and legal advertising. A telehealth provider, a financial services firm, a law practice, or an organization running issue or political messaging can expect more scrutiny of the advertiser and the specific creative before anything runs. The review exists because these categories carry regulatory and consumer-protection sensitivity that general retail does not.

If you operate in one of the restricted fields, plan for a more involved approval process and build campaigns that can stand up to it. If you operate in an excluded field, ChatGPT Ads is not currently a channel for you, and that is unlikely to change soon. Knowing which bucket you fall into is the first step, which is why we cover how to confirm eligibility at the end of this post.

How Does OpenAI Handle Brand Safety?

OpenAI runs brand safety on two tracks at once: who can advertise, and where ads are allowed to appear.

The first track is the category and advertiser review described above. Excluded categories are kept off the platform, and restricted categories pass through tighter scrutiny before approval. This screens advertisers and creative on the front end.

The second track protects the integrity of the answer itself. Ads appear below the relevant response, clearly labeled and separated from the answer, and the answers stay independent of advertiser influence. In other words, paying for placement does not buy a more favorable answer. The model responds first, and any eligible ad sits beneath that response as a distinct unit.

There is also a privacy layer that supports brand safety. Conversations stay private from advertisers, so you target topics and contexts, not identified individuals or their chat content. That structure limits how far advertising can intrude on a user's private exchange with the assistant. Together, category review, clear ad separation, and conversation privacy form the brand safety framework that governs the platform today.

What Sensitive Conversation Contexts Are Off-Limits?

Category eligibility decides whether you can advertise. Context decides whether any ad runs in a given moment, and some conversations are off-limits regardless of advertiser.

OpenAI keeps ads out of sensitive contexts, specifically personal health, mental health, and politics. If a user is discussing a medical concern, working through a mental health issue, or engaged in a political conversation, that exchange is treated as a space where advertising does not belong. The exclusion applies to the context, not just the advertiser, so even an eligible business will not appear there.

This matters for two reasons. First, it explains why a perfectly eligible advertiser can be absent from conversations that seem topically related. A wellness brand will not surface inside a user's discussion of a personal health scare, by design. Second, it sets the tone for the platform: ads are positioned as helpful commercial suggestions in appropriate moments, not interruptions during vulnerable ones.

For advertisers, the lesson is to think in terms of relevant and appropriate commercial intent rather than raw topic adjacency. The system is built to protect the user experience in sensitive moments, and that protection sits above any targeting you set.

What Brand-Safety Controls Do Advertisers Have?

The strongest brand-safety controls on ChatGPT Ads are the inputs you give the system, because contextual relevance is what drives placement.

Your context hints, landing page, ad title, and ad copy all shape where your ads can appear. The system selects primarily on relevance to the live conversation, so tightly defined hints and clear creative steer you toward the contexts you actually want and away from loosely related ones. Sharpening these inputs is the closest equivalent to traditional placement and keyword controls, and it is where your influence is greatest.

You also have measurement controls. ChatGPT Ads support the OpenAI pixel, the Conversions API, and UTMs, with reporting on impressions, clicks, spend, click-through rate, average cost per click, average cost per mille, and conversions. That visibility lets you see which contexts produce qualified outcomes and refine accordingly.

What you do not get is a window into private conversations or the ability to target named individuals. Conversations stay private from advertisers, which is a deliberate guardrail rather than a missing feature. Your control operates at the level of relevance and creative, not surveillance, and that is the model to plan around. For the creative side of that work, see our guide to writing ad creative for ChatGPT.

What Should Regulated Businesses Know Before Advertising?

If you operate in health, financial, political, or legal services, treat ChatGPT Ads as a channel you can use, with conditions, rather than one to assume.

Expect review at the advertiser, ad, and placement level. That means your business is assessed, your individual ads are assessed, and the contexts where they could appear are assessed. Build creative that is accurate, compliant, and free of claims that would draw scrutiny, because restricted categories pass through tighter review before anything runs. Vague or aggressive messaging is more likely to stall in approval.

Also plan around the sensitive-context exclusions. A healthcare advertiser is eligible in principle, yet ads are kept out of personal health and mental health conversations, so your reachable inventory is narrower than the topic volume might suggest. The same logic applies to political messaging given the politics exclusion. Model your expectations on appropriate commercial moments, not on every conversation that mentions your field.

The platform is still expanding and the rules are evolving. We have early hands-on experience with ChatGPT Ads, including direct work with OpenAI's activation team, and the consistent pattern is that regulated advertisers succeed by leading with compliant, specific, genuinely useful offers. If you want broader context first, start with the complete guide to ChatGPT Ads.

How Do You Confirm Your Category Is Eligible?

Confirming eligibility is a short process, and it is worth doing before you invest in building campaigns.

Start with the category test. If your business is mainstream and would run without issue on Google or Meta, you are very likely eligible. If you fall into adult, alcohol, tobacco, or gambling, you are excluded. If you fall into health, financial, political, or legal, you are restricted and should plan for tighter review. This single sort tells most advertisers what they need to know.

Next, check the access basics, because eligibility also depends on the platform. Self-serve advertising opened to U.S. businesses through ads.openai.com in May 2026, with no minimum spend, and ads serve to logged-in Free and Go tier users in the United States, Canada, Australia, and New Zealand, plus the United Kingdom as of June 6, 2026 in an early-stage, registration-gated rollout, with under-18 users excluded. Confirm your audience actually sits inside that footprint before committing budget. Our overview of how to advertise on ChatGPT and our breakdown of who sees ChatGPT Ads cover both points in detail.

Finally, when you are ready, the definitive confirmation comes from submitting your business and creative through the Ads Manager review. Approval there is the real answer, especially for restricted categories. If you would rather have that assessed for you, we can review your category fit and brand-safety exposure as part of a free marketing audit.