In a single week, three of the most consequential platforms in eCommerce advertising each made a move that, taken individually, would be worth paying attention to. Taken together, they amount to a structural shift: the path from product discovery to purchase is being rebuilt around AI, and every major platform is racing to own a piece of that journey.
OpenAI is expanding ChatGPT ads to five new markets and testing multi-advertiser placements. Amazon has launched an AI image generator directly inside its shopping app search bar. Google has rolled out new Performance Max asset experimentation tools that finally give advertisers controlled creative testing. The question for eCommerce merchants is no longer whether AI is reshaping advertising. It is how quickly you adapt your strategy to the environment that already exists.
The ChatGPT advertising expansion
When OpenAI first introduced advertising in ChatGPT in February 2026, it was a limited experiment in the US. The programme has since expanded to Canada, Australia, and New Zealand, and as reported by Search Engine Land, OpenAI is now rolling out to five additional markets, including the UK, Japan, South Korea, Brazil, and Mexico.
More significantly, OpenAI is testing multi-advertiser placements, moving from a single-sponsor model to a competitive auction format sold through a second-price auction mechanism. A self-serve Ads Manager beta launched on 5 May with cost-per-click bidding alongside the existing CPM model, with recommended starting bids of $3 to $5 per click. The ad experiment crossed $100 million in annualised revenue within six weeks of launching, according to an OpenAI spokesperson quoted by Reuters.
For eCommerce merchants, this creates a clear opportunity. ChatGPT is increasingly where consumers begin product research. When someone asks "what is the best running shoe for flat feet" or "which espresso machine under £500 should I buy," they are expressing purchase intent in a format that is far more qualified than a traditional search query. Advertising in that context means reaching consumers at a moment of high intent with extremely specific product needs.
The challenge is that this is yet another advertising channel to manage, optimise, and budget for. Unlike Google or Meta advertising, where decades of institutional knowledge exist, ChatGPT advertising is genuinely new territory with limited performance benchmarks. Ads appear only for logged-in users on the Free and Go tiers, meaning paid subscribers remain ad-free.
Amazon's visual AI search: Raising the bar
While OpenAI builds its ad business, Amazon is approaching the same problem from the opposite direction, using AI to make its existing marketplace more discoverable. As announced by Amazon on 3 June and reported by Retail Dive, a new AI image generator inside the Amazon Shopping app creates images in real time as customers type descriptive queries. The feature generates visual suggestions based on criteria such as pattern, colour, and texture, refining with each additional word.
This is a fundamental shift in how product search works. Instead of typing keywords and scrolling through results, shoppers describe what they want in natural language and visually confirm their intent before seeing any products. The feature currently covers apparel and home goods, with plans to expand across additional categories. It launched alongside five other visual shopping features, including Shop by Style (AI-generated shoppable outfit collages), Amazon Lens Live (real-time camera search), and the integration of Alexa for Shopping directly into the main search bar, which replaced the Rufus chatbot in May 2026.
For merchants selling on Amazon, the implication is direct: your product images and attributes need to be rich enough to match against AI-generated visual queries. Traditional keyword-optimised listings may not surface when the discovery mechanism is visual rather than textual. Product photography, detailed attribute tagging, and visual consistency across your catalogue become even more critical.
Google's Performance Max gets smarter testing
Google is not standing still either. As reported by Search Engine Land on 8 June, Google is rolling out new asset experiments for Performance Max campaigns, allowing advertisers to compare entirely new asset groups, evaluate the impact of adding individual assets, or measure the performance of seasonal creative against evergreen content.
This matters because Performance Max campaigns have historically operated as something of a black box. You feed in assets, and Google's AI decides how to combine and deploy them. The new experimentation tools represent a shift towards transparency, enabling advertisers to run controlled A/B comparisons of creative inside a single campaign for the first time. According to Google's internal data, advertisers using the asset testing feature are seeing an average of 14% more conversions compared to campaigns without structured asset testing.
For eCommerce merchants running PMax campaigns, this means you can finally distinguish between a high-performing product image and one that is dragging down your campaign, and act accordingly.
What connects all three
What links these developments is a single underlying trend: AI is restructuring the entire path from discovery to purchase, and every major platform is racing to own its share of that journey.
Consider a typical product purchase path in mid-2026. A consumer asks ChatGPT for product recommendations, where they now encounter ads. They search on Amazon using natural language and visual AI tools. They encounter Performance Max ads across Google's properties with AI-optimised creative. Each of these touchpoints is mediated by AI models making real-time decisions about what to show, to whom, and when.
For merchants, this means the old model of "get your SEO right and run some Google Ads" is no longer sufficient. You need a presence in AI recommendation engines, visual search databases, and conversational advertising platforms simultaneously.
Practical steps for eCommerce merchants
Audit your product content for AI readability
Your product descriptions, attributes, and imagery need to work for machines, not just humans. Structured data, comprehensive attribute tagging, and high-quality photography from multiple angles are table stakes. Amazon's visual AI search makes the quality gap between merchants with professional photography and those using manufacturer stock images even more consequential.
Start tracking AI visibility now
Monitor brand mention frequency in AI responses, track direct traffic trends (which increasingly mask AI-referred visits), and test how your products appear in ChatGPT, Google AI Overview, and Amazon's visual search. The metrics that matter most are still evolving, but establishing a baseline now means you will have comparative data when the landscape matures.
Budget for experimentation
If you are not already testing ChatGPT ads, the expansion to new markets is a natural entry point. Early movers in any advertising channel typically benefit from lower competition and CPCs before the market matures. The recommended starting CPC of $3 to $5 is competitive with mature paid search in many eCommerce verticals.
Invest in visual assets
Amazon's visual AI search makes professional product photography a competitive differentiator, not a nice-to-have. If your competitors have better product imagery, they will increasingly win in AI-mediated discovery. Consider investing in lifestyle photography, 360-degree product views, and consistent visual language across your catalogue.
Use the new PMax testing tools
Google's new asset experiments are not optional. They are a competitive advantage. Use them to systematically identify your highest-performing creative elements and allocate budget accordingly. The 14% conversion lift that Google reports for structured asset testing is too significant to leave on the table.
The bigger picture
The advertising landscape is fragmenting across AI platforms at a speed that few predicted even 12 months ago. The merchants who succeed will be those who recognise that AI is not just changing how ads are served. It is changing how products are discovered, evaluated, and purchased. Every touchpoint in the customer journey is being rebuilt with AI at its core, and your marketing strategy needs to reflect that reality.
The window for early-mover advantage in AI advertising is narrowing. The expansion of ChatGPT ads, the sophistication of Amazon's visual search, and the improved transparency in Performance Max all point to a market that is maturing rapidly. The time to build your AI advertising competency is now, while there is still room to experiment and learn before the stakes get higher.
About On Tap
On Tap is a growth-focused eCommerce consultancy helping mid-market and enterprise merchants navigate the intersection of advertising technology, product content strategy, and platform optimisation. From AI readiness audits and product feed optimisation to paid media strategy across emerging channels, On Tap works with merchants to build advertising capabilities that compound over time.
If you want to understand where AI advertising fits in your growth strategy, get in touch.
