1d:I[70616,["/_next/static/chunks/0kcf5dt6yxdox.js","/_next/static/chunks/1su5n1uv1ti4s.js","/_next/static/chunks/0gvke1wv6udfg.js","/_next/static/chunks/1dfrle3jgmlzh.js","/_next/static/chunks/162lgvduyr-fa.js","/_next/static/chunks/2gcnnopam8e9w.js","/_next/static/chunks/2g60iab5a1pkg.js"],"Breadcrumb"] 1e:I[90762,["/_next/static/chunks/0kcf5dt6yxdox.js","/_next/static/chunks/1su5n1uv1ti4s.js","/_next/static/chunks/0gvke1wv6udfg.js","/_next/static/chunks/1dfrle3jgmlzh.js","/_next/static/chunks/162lgvduyr-fa.js","/_next/static/chunks/2gcnnopam8e9w.js","/_next/static/chunks/2g60iab5a1pkg.js"],"ArticleGrid"] 1f:T1933,

Asked which phone to buy, Google's AI Overviews reports that a particular model is widely rated best. ChatGPT says: my pick right now. That difference in voice is not a stylistic quirk, according to an audit published this week — it is a measurable and systematic split in how AI assistants deliver commercial advice, arriving just as both companies build advertising into the same surfaces.

Key takeaways

Across 1,536 evaluated responses, ChatGPT expressed a first-person product preference in 79% of product-recommending answers, against 7% for Gemini and 2% for Google AI Overviews.
For the same query, the ChatGPT and Gemini interfaces displayed an average of just 5.4% of the same source domains, and shared no domain at all in 76.7% of comparisons.
Provider APIs did not reproduce what consumers see, overlapping their own interfaces on only 12.0% of domains for ChatGPT and 14.8% for Gemini — which undercuts the standard method for auditing these systems.

How the audit was built

The study, posted to arXiv by a team from Maastricht University, Utrecht University and the University of Zurich, starts from a dataset the authors call ConsumerQ: 2,528 commercial-advice queries written by real users rather than by researchers. From that pool they drew 117 genuine product-recommendation questions and ran each through five surfaces — the ChatGPT and Gemini consumer interfaces, both of their APIs, and Google Search's AI Overviews — for 1,755 observations collected from a fixed location, with every interface request paired to an API request.

Pairing matters because the paper's sharpest finding is about method rather than about any one assistant. Most published audits of generative search run through provider APIs using researcher-written template queries, for the obvious reasons of cost and reproducibility. The authors argue that for commercial advice specifically, that shortcut measures something consumers never encounter.

Three assistants, three different kinds of answer

The framing gap is the most visible result. ChatGPT states a preference in the first person in 79% of responses, against 7% for Gemini and 2% for AI Overviews. It also labels some product the best in 73% of responses, where Gemini does so 43% of the time and AI Overviews 48%. The same question therefore arrives as a personal recommendation, an undisputed winner, or a menu of options, depending only on which product the shopper opened.

None of it is stable. The recommended products change across repeated identical requests, and even the framing flips — a system presents a product as the best in 27 to 40% of repeat cases where it had not done so before. A shopper who asks twice is not asking the same question twice.

The citation layer barely overlaps

Sources diverge further than the answers do. For an identical query, the ChatGPT and Gemini interfaces share on average 5.4% of displayed domains, and in 76.7% of comparisons they have no domain in common whatsoever. Repeated requests keep surfacing new domains rather than converging on a stable set.

That spread has a commercial explanation. Which publishers an assistant can draw on depends on licensing: OpenAI and Google both hold agreements for real-time Reddit data, and OpenAI's deal with Axel Springer puts Politico and Business Insider into OpenAI's pool. The retrieval layer is not a neutral index — it is a negotiated one.

Why regulators should care about the API gap

The API result is the one with teeth. Mean domain overlap between an interface and its own API runs 12.0% for ChatGPT and 14.8% for Gemini, and the two expose different types and layers of source information entirely. An auditor querying the API is looking at a different system from the one a consumer uses.

The timing is pointed. On 31 August 2026 the European Commission designated ChatGPT a Very Large Online Search Engine under the Digital Services Act, which brings obligations on systemic risk — including consumer protection in online purchasing — plus independent audits and researcher data access. The paper's conclusion is that such audits cannot be run through the API and still claim to describe consumer experience.

Commercial pressure is building on the same surface. OpenAI expanded advertising in ChatGPT to 31 European markets in August, and product-recommendation queries already account for roughly 2% of ChatGPT conversations by the largest published usage measurement. We have covered OpenAI's experiments with conversational ad formats; this audit supplies the baseline against which any later change in recommendation behaviour can be measured.

The authors' recommendation to fellow researchers is procedural rather than alarmist: audit the consumer interface, repeat every query, and state which source layer you observed. Single responses collected through an API cannot stand in for the advice people actually receive from a large language model at the point of purchase.

FAQ

Does the study show ChatGPT recommends products for money?

No. The audit measures how recommendations are framed, how consistent they are, and which sources appear — not whether any placement was paid. Its contribution is a documented pre-advertising baseline and a method that would detect a shift, which the authors argue is what regulators and researchers currently lack.

Why can't researchers just use the ChatGPT and Gemini APIs?

Because the APIs return a materially different view. Mean domain overlap with the corresponding consumer interface was 12.0% for ChatGPT and 14.8% for Gemini, and the source information the two layers expose differs in kind, so API observations cannot be assumed to represent what shoppers see.

Is the query dataset available?

The authors curated ConsumerQ, a set of 2,528 real commercial-advice queries written by users, and drew the 117 product-recommendation questions analysed in the paper from it. Full details of the dataset and the collection protocol are in the arXiv preprint.

a:["$","div",null,{"className":"mx-auto max-w-7xl px-4 py-8 sm:px-6","children":[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"BreadcrumbList\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https://ainewsway.com\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"#google-gemini\",\"item\":\"https://ainewsway.com/tag/google-gemini\"}]}"}}],["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"ItemList\",\"@id\":\"https://ainewsway.com/tag/google-gemini#itemlist\",\"name\":\"#google-gemini Articles\",\"numberOfItems\":1,\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"url\":\"https://ainewsway.com/articles/chatgpt-picks-a-favourite-in-79-of-shopping-answers-gemini-does-it-7-of-the-time-mu5b5cih\",\"name\":\"ChatGPT Picks a Favourite in 79% of Shopping Answers. Gemini Does It 7% of the Time.\"}]}"}}],["$","$L1d",null,{"items":[{"label":"# google-gemini"}]}],["$","h1",null,{"className":"mb-8 text-3xl font-bold text-zinc-900 dark:text-zinc-100","children":["#","google-gemini"]}],[["$","$L1e",null,{"articles":[{"id":"8d40e102-d232-44c6-b542-36768ebe051b","slug":"chatgpt-picks-a-favourite-in-79-of-shopping-answers-gemini-does-it-7-of-the-time-mu5b5cih","title":"ChatGPT Picks a Favourite in 79% of Shopping Answers. Gemini Does It 7% of the Time.","subtitle":"An audit of 1,536 responses finds the two assistants barely cite the same sources — and that API-based research misses what shoppers actually see","content":"$1f","excerpt":"An audit of 1,536 AI shopping answers found ChatGPT states a first-person preference 79% of the time, Gemini 7% — and their cited sources barely overlap.","coverImage":"https://cdn.ainewsway.com/uploads/2026/09/c465dfe5-929c-4e81-a26f-e07ed45c6341.webp","coverImageCaption":"Online shopping illustration — the audit covered 117 real product-recommendation queries put to ChatGPT, Gemini and Google AI Overviews.","authorId":"6773f67d-d7fa-4b19-8f38-384f10ac761b","status":"PUBLISHED","publishedAt":"2026-09-17T09:09:22.132Z","viewCount":0,"likeCount":0,"readingTime":5,"isFeatured":false,"isBreakingNews":false,"metaTitle":"ChatGPT Names a Favourite in 79% of Shopping Answers","metaDescription":"An audit of 1,536 responses found ChatGPT states a first-person product pick 79% of the time vs Gemini's 7% — and their cited sources rarely overlap.","metaKeywords":["AI product recommendations","ChatGPT shopping","Gemini","AI Overviews","AI audit","Digital Services Act","AI advertising"],"ogImage":null,"aiSummary":"An academic audit of 1,536 AI chatbot responses found ChatGPT expresses a first-person product preference in 79% of shopping answers, compared with 7% for Gemini and 2% for Google AI Overviews. The interfaces shared only 5.4% of cited source domains on average. Provider APIs overlapped their own consumer interfaces on just 12 to 15% of domains, meaning API-based audits do not capture the commercial advice shoppers actually receive.","aiTags":[],"sentiment":null,"articleType":"ANALYSIS","videoEmbed":null,"videoType":null,"seoScore":85,"isAutoGenerated":true,"isBackfill":false,"pipelineStage":null,"copyrightScore":null,"similarityScore":null,"matchedSourceId":null,"createdAt":"2026-09-17T09:08:56.346Z","updatedAt":"2026-09-17T12:03:31.139Z","contentUpdatedAt":"2026-09-17T12:03:31.139Z","deletedAt":null,"author":{"id":"6773f67d-d7fa-4b19-8f38-384f10ac761b","name":"Seung Jung","username":"Seung Jung","avatar":null},"categories":[{"id":"b84a09d8-4faa-4fea-8c60-a37c2cf9f593","name":"LLM & Chatbots","slug":"llm-chatbot"}],"tags":[{"id":"c7c539a8-3b82-4395-98ce-1a285d235daf","name":"ChatGPT","slug":"chatgpt"},{"id":"166ded85-8723-4245-bd73-57941512cc9b","name":"Google Gemini","slug":"google-gemini"},{"id":"a095a81a-7a96-40cd-88de-019282d1efd0","name":"AI Shopping","slug":"ai-shopping"},{"id":"8c0ab43a-cf14-44f2-89ae-d05105177e65","name":"AI Audit","slug":"ai-audit"},{"id":"9b363558-6673-4fbe-962a-a4579018675b","name":"Digital Services Act","slug":"digital-services-act"}],"translations":[{"id":"560833ea-a24d-4d90-b878-6b28a068e030","language":"ja","title":"ChatGPTは買い物の回答の79%で「推し」を名指しする、Geminiは7%","subtitle":"1,536件の応答を監査した結果、両アシスタントが引用する情報源はほとんど重ならなかった。API経由の調査は消費者が実際に見る画面を捉えていない","excerpt":"AIの買い物回答1,536件を監査したところ、ChatGPTは79%で一人称の好みを示したのに対しGeminiは7%。引用元もほとんど重ならなかった。","metaTitle":"ChatGPT、買い物回答の79%で特定商品を名指し推薦","metaDescription":"AIの買い物回答1,536件の監査で、ChatGPTは79%が一人称の推薦、Geminiは7%だった。両者が表示する引用ドメインの一致は平均5.4%にとどまる。"},{"id":"020ec48c-9383-410d-958a-ae3873640000","language":"ko","title":"ChatGPT는 쇼핑 답변 79%에서 콕 집어 추천한다, Gemini는 7%","subtitle":"응답 1,536건을 감사한 결과 두 어시스턴트가 인용한 출처는 거의 겹치지 않았다 — API로 하는 연구는 소비자가 실제로 보는 화면을 놓친다","excerpt":"AI 쇼핑 답변 1,536건 감사에서 ChatGPT는 79%가 1인칭 선호를 밝힌 반면 Gemini는 7%에 그쳤다. 인용 출처도 거의 겹치지 않았다.","metaTitle":"ChatGPT, 쇼핑 답변 79%에서 특정 제품 콕 집어 추천","metaDescription":"AI 쇼핑 답변 1,536건 감사 결과 ChatGPT는 79%에서 1인칭 추천을 내놨고 Gemini는 7%였다. 두 어시스턴트의 인용 출처는 평균 5.4%만 겹쳤다."}],"celebrities":[{"id":"b086135e-3126-46cd-b591-67f073dff1bb","nameKo":"OpenAI","nameEn":"OpenAI","slug":"openai","profileImage":"https://ui-avatars.com/api/?name=OpenAI&size=400&background=F97316&color=fff&bold=true"},{"id":"5c572908-2f5b-47b3-95e2-13ed3e298bf7","nameKo":"구글","nameEn":"Google","slug":"google","profileImage":null}]}]}],"$L20"]]}]

#google-gemini

ChatGPT는 쇼핑 답변 79%에서 콕 집어 추천한다, Gemini는 7%

뉴스레터 구독

#google-gemini

ChatGPT는 쇼핑 답변 79%에서 콕 집어 추천한다, Gemini는 7%