OpenAI Launches ChatGPT Ads Manager Plugin, Requires Product Feeds for Shopping Results, Expands to New Regions
OpenAI announced significant updates to ChatGPT's advertising platform on September 4, 2026, introducing a plugin that lets advertisers manage campaigns directly from ChatGPT or Codex, requiring product feeds for shopping results, and expanding ad availability to Europe, India, the Middle East, and North Africa.
What was announced
OpenAI rolled out a comprehensive set of updates to ChatGPT Ads, headlined by the launch of the ChatGPT Ads Manager plugin and a fundamental change to how shopping results work: only brands that submit product feeds now appear in ChatGPT's shopping results. The updates also include custom audience improvements, new conversion matching fields, carousel ad insights, budget pacing for total-budget campaigns, and geographic expansion to select countries across four regions.
Background and context
ChatGPT's advertising platform has evolved rapidly since its initial launch. OpenAI introduced ads to ChatGPT's free tier in early 2026, initially testing the format with limited advertiser access. By July 2026, the platform had gained conversion tracking, audience targeting controls, and a dedicated Ads Manager interface. The August 2026 expansion brought ads to 31 European countries, signaling OpenAI's commitment to scaling the platform globally.
The September 2026 updates represent a maturation phase. The product feed requirement for shopping results aligns ChatGPT with traditional e-commerce search engines like Google Shopping and Amazon, where structured product data is mandatory for visibility. The Ads Manager plugin goes further than any competing AI search platform has attempted, letting advertisers use natural language to create, modify, and troubleshoot campaigns without leaving the ChatGPT interface.
What changed — specific details
Product feeds now required for shopping results
The most impactful change for brands is that ChatGPT shopping results now exclusively pull from submitted product feeds. Analyst Juozas Kaziukėnas confirmed on LinkedIn that brands without product feeds no longer appear in shopping results. This means:
- Brands must submit structured product data to OpenAI to appear in shopping results
- Product information (titles, descriptions, prices, images, availability) must be provided in a machine-readable format
- Brands that relied on organic mentions or unstructured product data in ChatGPT responses will lose shopping visibility
This change fundamentally shifts ChatGPT shopping from an algorithmic recommendation system to a structured commerce platform, similar to how Google Shopping requires Merchant Center feeds.
ChatGPT Ads Manager plugin
OpenAI launched a plugin that integrates campaign management directly into ChatGPT and Codex. Advertisers can:
- Use plain language to turn a website or brief into ready-to-launch ads
- Create ad variants and update campaigns through conversation
- Troubleshoot delivery issues and review performance metrics
- Receive prioritized recommendations for optimization
The plugin previews recommended changes and requires confirmation before applying them, addressing concerns about automated changes to live campaigns. To access it, advertisers add the "ChatGPT Ads Manager" from the Plugins section and connect their account.
Custom audience improvements
OpenAI removed several friction points from audience management:
- Advertisers can add, remove, or replace audience members without recreating the entire audience
- Multiple identifier types can be mixed in the same request
- Audiences below 25,000 matched users can now be used for exclusions (previously restricted)
- Audience size limits expanded beyond 5 million members
- More granular audience-size ranges are now visible
- Google Advertising ID (GAID) added as a supported identifier type for mobile targeting
New conversion matching fields
OpenAI expanded the data fields available for conversion tracking across both the Measurement Pixel and Conversions API:
Pixel now supports:
- Hashed phone numbers
- First name and last name (hashed)
- Region and postal code
Conversions API now supports:
- Plural list fields for hashed email, phone, external ID, first and last name
- Country, city, region, and postal code
- Android GAID for mobile conversion tracking
These additions improve match rates and attribution accuracy, particularly for mobile and cross-device campaigns.
Carousel ad unit format insights
For campaigns using the multi-product carousel ad format, reporting now includes:
- Carousel card impressions (counted when an individual product card becomes viewable)
- Carousel card clicks
- Product-level fields in the public Insights API
Carousel card impressions are separate from billable ad impressions and are not included in billable impression counts. Complete data is available from August 20, 2026, on a rolling 30-day basis.
Campaign total budget pacing
Campaigns with total budgets (rather than daily budgets) now have pacing to help spread spend more evenly over the campaign duration. OpenAI notes that actual spend may still vary day to day based on available delivery opportunities.
Conversion optimization options coming soon
OpenAI announced a new conversion optimization model that will incorporate view-through conversions. This model optimizes based on conversions from both ad clicks and ad views. With this campaign type, billing remains impression-based while delivery optimizes toward the selected conversion goal. OpenAI has not shared a specific rollout timeline.
Geographic expansion
ChatGPT Ads are now live in select countries across Europe, India, the Middle East, and North Africa. This expansion follows the August 2026 rollout to 31 European countries and extends availability to new markets. Advertisers with business in these regions can now serve ads to ChatGPT users.
Impact and implications
For brands and advertisers
The product feed requirement is the most significant operational change. Brands that appeared in ChatGPT shopping results through organic mentions or unstructured data must now submit product feeds to maintain visibility. This creates a new dependency on structured data submission, similar to Google Merchant Center or Amazon's catalog requirements.
Action items:
- Audit current ChatGPT shopping visibility to identify which products may lose placement
- Prepare product feed infrastructure if not already in place
- Submit product feeds to OpenAI with complete, accurate product data
- Monitor shopping result placement after the transition
The Ads Manager plugin reduces the technical barrier to campaign management. Advertisers without dedicated programmatic expertise can now use natural language to create and optimize campaigns. However, this also means OpenAI has more control over campaign structure and may push advertisers toward specific formats or strategies through the plugin's recommendations.
The audience improvements remove previous limitations that constrained targeting flexibility. The ability to use smaller audiences for exclusions and mix identifier types gives advertisers more granular control over who sees their ads.
For SEO and GEO practitioners
The product feed requirement signals that AI search platforms are moving toward structured commerce data as the foundation for shopping results. This has several implications:
Product data becomes critical for AI search visibility. Just as technical SEO (site speed, mobile-friendliness, structured data) became essential for Google Search, structured product data is now essential for AI search shopping results. Brands that neglect product feed optimization will lose visibility in ChatGPT and likely other AI search platforms that follow this model.
The distinction between organic AI citations and paid shopping results becomes clearer. Previously, brands could appear in ChatGPT shopping results through organic mentions in training data or real-time web search. Now, shopping results are explicitly tied to paid product feeds, while organic citations remain separate. This creates a clearer division between earned AI visibility (through content quality and brand authority) and paid AI visibility (through product feeds and ads).
Multi-platform product feed management becomes a core competency. Brands now need to maintain product feeds for Google Merchant Center, Amazon, Microsoft Advertising, and OpenAI. Each platform may have different format requirements, update frequencies, and data quality standards. Centralizing product data management and automating feed distribution across platforms will become increasingly important.
For the broader AI search ecosystem
OpenAI's product feed requirement may set a precedent for other AI search platforms. If ChatGPT's shopping results prove more effective with structured product data, competitors like Google AI Mode, Microsoft Copilot, and Perplexity may adopt similar requirements. This could accelerate the shift from algorithmic product recommendations to structured commerce data as the foundation for AI search shopping.
The Ads Manager plugin also represents a new approach to advertiser tooling. By embedding campaign management in the conversational interface, OpenAI is testing whether natural language is a more intuitive way to manage ads than traditional dashboards. If this approach succeeds, it could influence how other platforms design their advertiser interfaces.
Industry reaction and expert perspectives
Juozas Kaziukėnas, an analyst who tracks AI search developments, confirmed the product feed requirement on LinkedIn, noting that shopping results now exclusively come from brands that submit product feeds. This confirmation from an independent analyst adds credibility to the change and suggests it's a deliberate platform strategy rather than a temporary testing phase.
The broader industry reaction has focused on the implications for brands that built AI search visibility through organic means. Some practitioners question whether the product feed requirement will reduce the diversity of products shown in shopping results, as smaller brands without feed infrastructure may lose visibility. Others see it as a necessary step to improve result quality and advertiser accountability.
What has not been confirmed
Several aspects of the September 2026 updates remain unclear:
Product feed format requirements. OpenAI has not published detailed specifications for the required product feed format. It's unclear whether OpenAI will accept standard formats like Google Merchant Center feeds or require a proprietary format.
Timeline for full enforcement. The product feed requirement appears to be in effect, but the transition timeline for brands currently appearing in shopping results without feeds is not specified.
Pricing and auction mechanics. OpenAI has not shared details on how product feed submissions interact with the ad auction. It's unclear whether submitting a feed guarantees placement or whether feeds enter a competitive auction based on bid amounts and relevance.
View-through conversion optimization rollout. The announced conversion optimization model that includes view-through conversions has no confirmed rollout date.
Impact on organic AI citations. It's unclear whether the product feed requirement affects organic AI citations in non-shopping contexts. Brands that appeared in ChatGPT responses through organic mentions may still retain that visibility, but the relationship between paid shopping results and organic citations needs clarification.
What to watch next
Competitor responses. Watch for Google, Microsoft, and Perplexity to announce similar product feed requirements for their AI search shopping results. If multiple platforms adopt this model, structured product data management becomes a universal requirement for AI search visibility.
Feed format standardization. If OpenAI, Google, and Microsoft require different feed formats, the industry may push for a standardized format to reduce complexity. Alternatively, feed management platforms may emerge to handle multi-platform distribution.
Plugin adoption and effectiveness. Monitor advertiser adoption of the ChatGPT Ads Manager plugin and whether natural language campaign management proves more effective than traditional dashboards. If adoption is high, other platforms may follow OpenAI's approach.
Shopping result quality. Track whether the product feed requirement improves the relevance and quality of shopping results in ChatGPT. If result quality improves significantly, it could accelerate the shift of shopping queries from traditional search to AI search platforms.
Regulatory scrutiny. As AI search platforms expand their advertising capabilities, they may face increased regulatory scrutiny around data collection, targeting practices, and disclosure of paid placements. The expansion to Europe, India, the Middle East, and North Africa brings ChatGPT Ads under multiple regulatory regimes with different requirements.
Sources
- OpenAI official email to advertisers (September 4, 2026)
- Juozas Kaziukėnas, LinkedIn post confirming product feed requirement for ChatGPT shopping results (September 2026)
- OpenAI ChatGPT Ads Manager plugin documentation
- OpenAI custom audience improvements documentation
- OpenAI Measurement Pixel and Conversions API documentation
