1. Scope and status
An agency policy, not an OpenAI policy.
Advertise for AI is an advertising and marketing consultancy. We are not affiliated with, endorsed by or sponsored by OpenAI. This page explains which engagements we will accept, how we review campaign materials and when we will decline, pause or require changes.
OpenAI may change its policies, supported markets, advertiser access, category eligibility, review requirements and delivery systems. The current official OpenAI policy and platform decision prevail over this page in every case.
Open official policy ↗2. Baseline standards
Every asset must be truthful, professional and clearly advertising.
No misleading claims
Capabilities, prices, outcomes, affiliations, endorsements and comparisons must be accurate, current and supportable.
No interface imitation
Creative must remain visibly distinct from ChatGPT and other OpenAI product interfaces, voices and system messages.
No offensive or discriminatory content
Obscene, shocking, derogatory, harassing, defamatory or exclusionary content is declined.
No destination bait-and-switch
The landing page must match the advertiser, offer, claim, category, price and next action presented in the ad.
3. Preferred categories
Our primary intake focus follows the currently supported consumer categories.
Lifestyle, household and consumer goods
Retail goods, home products, apparel in a standard retail context, personal technology and practical consumer products.
Local services
Non-regulated local services with transparent identity, pricing or enquiry conditions and a clear destination.
Travel and experiences
Accommodation, transport-related consumer services, destinations, activities and non-gambling entertainment.
Digital products and education
Software, digital tools, courses and educational materials that do not provide prohibited regulated services or deceptive outcomes.
Eligibility is not automatic. The advertiser, creative, destination, geography and full offer are reviewed together.
4. Restricted categories
Accepted only after documented eligibility and market review.
United States only where OpenAI approval is available, the advertiser is appropriately licensed and the product is within the currently eligible list. Credit repair, debt settlement, debt assistance and alternative investments such as bullion are declined. Financial services outside the United States are generally declined.
Case by caseUnited States only where OpenAI approval is available, the advertiser is appropriately licensed and the product or service is within an eligible category. Unsupported treatment claims, unsafe or unapproved products, high-risk procedures and vulnerability exploitation are declined.
Case by caseUnited States only, where the advertiser is licensed to practise law in the jurisdiction in which the ad is shown. General legal education may be considered where no legal service is offered. Legal services outside the United States are declined.
US / licensedIndividual job listings and individual housing rental or sale listings are declined. Listing platforms may be considered only where the ad and landing page do not reference a specific job or property listing.
Platform onlyMay be considered where no medical claim is made and no distress, body-image vulnerability or unsupported outcome is exploited. Diet pills, detox claims and unsupported supplement claims are declined.
Claim review5. Prohibited categories
We will not accept campaigns in these areas.
6. Placement safety
We do not design campaigns for sensitive or brand-unsafe adjacency.
OpenAI’s current placement policy excludes harmful, controversial and vulnerable contexts. We will not ask clients to target, exploit or optimise toward child-safety issues, cyber abuse, dangerous activities, fraud, graphic sexual or violent content, hate, illicit activity, misinformation, politics, privacy violations, self-harm, terrorism, weapons, emotionally reliant contexts, mental or personal health conversations, or sensitive user journeys.
We do not promise that an ad will appear beside a specific prompt, conversation or user. Placement is controlled by OpenAI’s systems, safeguards, auction and availability.
7. Advertiser identity and destinations
The business, offer and destination must be coherent and verifiable.
- Advertisers must provide accurate legal and trading identity, ownership, location and affiliation information.
- Only owned or authorised trademarks, logos, music, footage, endorsements and other intellectual property may be used.
- Required professional licences, registrations and certifications must be current and available for review.
- The destination must clearly relate to the advertiser and offer, be accessible for review, and disclose material terms.
- The destination may not introduce prohibited products, services or claims that were absent from the creative.
- Advertise for AI must not be represented as OpenAI, ChatGPT, an official reseller, or an endorsed OpenAI partner.
8. Data and measurement
Use only lawful, authorised and appropriate data.
- Audience data must be first-party data where supported and must not be sourced from data brokers or marketplaces.
- Required notices, consent, permissions and legal bases must be in place before data is provided to any ad tool.
- Sensitive or prohibited personal data must not be placed in audience names, event names, parameters, tags or campaign fields.
- Reporting data may be used only for permitted campaign measurement and must not be used to identify individuals or create unauthorised profiles.
- Invalid clicks, impressions, queries, conversions or other interactions must never be generated manually or automatically.
9. Review and enforcement
Review continues after approval.
Advertiser and category review
Identity, licensing, geography, primary business model and eligibility.
Creative and destination review
Claims, assets, disclosures, IP, policy, accessibility and continuity.
Monitoring and remediation
Feedback, policy changes, destination changes, anomalies and new risk signals.
OpenAI’s current review model evaluates the advertiser, the ad creative and landing page, and the placement context. Review uses automated systems with human oversight, and ads or destinations that cannot be evaluated may be ineligible to run. OpenAI may reject, limit or remove ads or advertisers, including where a campaign conflicts with its advertising principles, business interests or competitive position.
Advertise for AI may independently decline, require edits, pause work or terminate an engagement where materials are unlawful, unsafe, misleading, prohibited, unverifiable, inaccessible for review, or inconsistent with current platform policy.
10. Policy changes
This page will change as the platform changes.
We review the official OpenAI policy during intake and before launch. The September 2026 source update is reflected here, including the current legal-services treatment, housing/job-listing rules and OpenAI’s stated discretion to decline advertising. Material policy changes may require revised creative, a different market, additional verification, a pause, or cancellation of the proposed campaign.