Valentius Kryptix

Here’s the honest answer, and it’s more specific than the “restricted, not banned” framing you’ll see in most general coverage of this topic: globally, OpenAI treats healthcare, financial services, and legal services as restricted categories eligible for case-by-case approval — but as of OpenAI’s own policy, last updated August 31, 2026, that approval pathway is explicitly limited to advertisers running ads shown in the US. OpenAI’s policy states plainly that ads for financial services, health services, and legal services outside the US are generally prohibited. If you’re an Indian hospital, NBFC, insurer, fintech app, or law firm, the current, practical answer is that self-serve ChatGPT advertising for your core regulated service isn’t open to you yet — not that you need better paperwork to get approved.

Key takeaways:

• Healthcare, financial services, and legal services all currently route through a US-only case-by-case approval pathway. OpenAI’s own policy states ads for these categories are “generally prohibited” outside the US.
• This means most Indian healthcare, finance, and legal businesses cannot currently run self-serve ChatGPT ads for their core regulated service, regardless of licensing or documentation quality.
• Separately, and permanently: ads never appear near personal health, mental health conversations, or political content, for any advertiser, in any market — this rule doesn’t depend on category approval.
• OpenAI’s ad policy has been revised six times since March 2026, most recently four days before this article was written. Treat everything here as current-as-of-publish, not fixed.
• Adjacent, non-regulated products — budgeting apps, general financial education, fitness trackers without medical claims — are a different category entirely and may already be open under standard review, wherever you’re based.

The Three Types of Rules That Actually Apply

It helps to keep three separate questions apart, because conflating them is exactly how “restricted, not banned” gets misread as “you can apply today.”

Category eligibility asks whether your type of business can advertise at all. Most consumer categories — household goods, local services, travel, education, digital products — are open under standard review, everywhere. Healthcare, financial services, and legal services are restricted, with case-by-case approval currently scoped to the US only. A short list of categories (adult content, gambling, political advertising, and others covered below) are prohibited outright, in every market, regardless of licensing.

Advertiser eligibility asks whether your specific business qualifies within an otherwise-open category — proof of licensure, business legitimacy, geographic compliance with wherever you’re targeting ads. Creative and claims review is a separate check on top of that: even an approved advertiser’s specific ad copy, imagery, and landing page get evaluated on their own terms — an approved healthcare advertiser can still have an individual ad rejected for an unsupported claim, the same way a cleared advertiser doesn’t get a blank check for whatever creative they submit.

Placement eligibility asks whether a specific conversation is an appropriate context for any ad, from any approved advertiser. This one applies universally and permanently: personal health conversations, mental health conversations, and other emotionally vulnerable contexts are excluded from ad placement regardless of who the advertiser is or whether their category is approved.

A healthcare brand can clear the first two gates entirely and still never have an ad appear next to a user’s personal medical question — because that third gate isn’t about the advertiser at all.

A quick reference, since these three categories get lumped together in most coverage of this topic:

Why Healthcare, Finance, and Legal Are Treated Differently

These three categories carry a combination of regulatory complexity and potential for real harm that most advertising categories don’t: a misleading claim about a medical treatment, an investment product, or legal representation can cause direct financial or health harm in a way a misleading ad for a backpack generally can’t. OpenAI’s policy reflects that by requiring proof of licensure for approved advertisers in these categories where the pathway is open, and by keeping the pathway closed outside the US for now — likely because the compliance infrastructure needed to verify licensing and regulatory standing varies enormously market to market, and building that out for one market (the US) before expanding is a reasonably cautious way to approach a category with real downside if reviewed badly.

It’s worth knowing what the US pathway actually covers, since it’s the clearest signal of what might eventually open to other markets if OpenAI expands this the way it has expanded other parts of the platform. On the health side, OpenAI names consumer medical devices and wearables, dental services, health insurance, hospitals and urgent care, medical testing, and vision products and services as examples eligible for case-by-case US approval — while unsafe or unapproved products, unsupported treatment claims, experimental therapies, and high-risk procedures remain off-limits even within the US. On the financial side, the named examples include auto loans, credit cards, deposit accounts, insurance, investment services, mortgages, personal loans, and payment services — while credit repair, debt settlement, and certain alternative investments are excluded even there. None of this is confirmed to be coming to India on any timeline; it’s simply the shape of the category as it currently exists where it’s open at all.

For the broader picture on ChatGPT Ads in India beyond this specific compliance question, see our complete guide, and if you’re weighing this channel generally rather than for a restricted category specifically, our readiness framework covers the wider set of factors worth checking before committing budget.

What’s Actually Open to Indian Advertisers in These Spaces Right Now

This is the distinction most worth understanding if you’re in a nominally regulated industry, because it’s not all-or-nothing. OpenAI’s policy draws a line between the regulated service itself and adjacent, non-regulated tools that happen to sit in the same industry.

On the financial side: “financial tools that do not promote financial products or transactions are generally permitted” — the policy names budgeting apps, general financial software, and financial education content that doesn’t itself offer a financial service. A loan, a credit card, an investment product, or insurance sold directly is restricted outside the US. A budgeting app that helps someone track spending without selling them a financial product is a different category, open under standard review.

On the health side, the same split exists: “general health and wellness products that do not make medical claims may be permitted,” including fitness equipment, wearable devices, and menstrual products — while a device or service claiming to diagnose, treat, or prevent a medical condition falls under the restricted Health Services category. A fitness tracker marketed as “track your steps and sleep” is a different product, as far as this policy is concerned, from the same tracker marketed as “clinically validated to detect atrial fibrillation.”

If your business sells the regulated product or service directly, this distinction won’t open a path around the restriction — don’t market a loan as a “budgeting tool” to route around the policy, since that’s exactly the kind of category-boundary gaming a review process is built to catch. But if you genuinely have an adjacent, non-regulated product line, it’s worth checking whether that specific offering already qualifies under standard review rather than assuming your whole business is blocked because your industry is regulated.

Categories That Are Fully Prohibited, Everywhere, Regardless of Market

Worth distinguishing from the health/finance/legal situation above: some categories aren’t restricted-pending-expansion, they’re prohibited outright, in every market, for every advertiser, with no case-by-case path at all. OpenAI’s policy names adult content and dating services, gambling, political advertising, alcohol and tobacco, recreational drugs, counterfeit goods, scams and fraud, and individual job or housing listings (though platforms carrying multiple listings can still advertise, provided the ad and landing page don’t reference one specific listing). These aren’t waiting for a geographic rollout the way health, finance, and legal are — they’re closed by design.

Ads Never Appear Near Personal Health or Mental Health Conversations — Even Where a Category Is Approved

This is the placement-level rule referenced above, and it’s worth stating precisely because it’s permanent and doesn’t depend on anything else in this article changing. OpenAI’s policy excludes ads from what it calls sensitive user contexts — emotionally reliant conversations, mental and personal health conversations, and other sensitive user journeys — as a standing rule, separate from and in addition to category-level restrictions. Even a fully approved US healthcare advertiser, cleared through the case-by-case process, will not have their ad appear next to someone’s personal medical question. The category approval controls whether you can advertise at all; this rule controls where an approved ad can never show up, regardless. For the fuller picture on what data is and isn’t used around sensitive topics, see our privacy guide.

Why the Rules Keep Changing

OpenAI’s ad policy has gone through six versions since its initial publication in March 2026: an April refinement to placement rules, a May addition explaining review standards, a July rewrite of the financial and health sections specifically to clarify eligible categories and markets, an August clarification on housing and job listings, and a further August update permitting legal services ads within the US. That’s roughly one substantive revision every five to six weeks. If you’re evaluating this space for a regulated business, the practical implication is to treat any specific claim — including everything in this article — as accurate as of its stated date, not as a fixed rule to plan a year of marketing around. Recheck OpenAI’s policy page directly before finalising any campaign plan in this category, however recently you last looked.

What Indian Healthcare & Finance Brands Should Actually Do Right Now

Given all of the above, the useful next steps look different from “gather your documentation and apply.” First, check honestly whether your actual product is the restricted regulated service or an adjacent, non-regulated tool — the distinction covered above may mean part of your business already qualifies under standard review even if your core service doesn’t. Second, if your core business is squarely inside the restricted category, treat paid ChatGPT advertising as currently unavailable to you rather than pending approval — there’s nothing to apply for yet, so there’s no application to prepare. Third, put the effort that would have gone into a paid campaign into the unpaid side of AI visibility instead: being cited or recommended in a ChatGPT answer organically doesn’t require advertiser approval at all, and is worth pursuing regardless of when — or whether — the paid pathway opens to India. Fourth, monitor OpenAI’s policy page directly rather than waiting for news coverage to catch up; given the revision pace above, geographic expansion could arrive with little advance notice. If and when it does, the account setup and campaign process itself follows the same steps as any other category — our step-by-step setup guide covers that groundwork so you’re not starting from zero once the pathway actually opens.

One More Gate, for Whenever This Does Open to India

Everything above covers OpenAI’s platform policy. It’s worth being explicit about a separate question this article doesn’t answer and can’t: even once the case-by-case pathway extends beyond the US, OpenAI’s approval is a platform decision, not a determination that a specific ad complies with Indian law. RBI advertising norms for financial products, ASCI’s code for healthcare and wellness claims, and sector-specific rules around medical advertising all sit entirely outside what OpenAI is checking for. A campaign could clear OpenAI’s review and still need changes — or full legal sign-off — to run safely in India. Platform approval and regulatory compliance are two different tests, and clearing one is never evidence of clearing the other.

That distinction also shapes what a claim should look like once eligibility exists. “This treatment will resolve your condition” is a fundamentally different (and far riskier) claim than “learn about treatment options and speak with a qualified professional” — not because one sounds nicer, but because the first asserts a guaranteed outcome and the second doesn’t. The same logic applies to financial claims: describing a product’s features and eligibility criteria is a different exercise from promising a specific return. None of this is a substitute for an actual compliance review of your specific claims — it’s the kind of distinction worth having in mind before that review happens, not instead of it.

Not sure whether your specific product falls inside a restricted category or qualifies under standard review? Valentius Kryptix will assess your actual offering against OpenAI’s current policy before you spend time on a paid channel that might not be open to you yet — and help you build the organic AI-visibility side in the meantime. Talk to our team →

Frequently Asked Questions