Question
Why ChatGPT With Browse Cites Different Sources Than Google AI Overviews
Chat assistants 4 min read
ChatGPT and Google AI Overviews draw their sources from different places and favor different things. An observational study by Seer Interactive, published in February 2025, matched more than 500 ChatGPT search citations against search results and found that 87% of SearchGPT citations matched Bing’s top results. Google, for its part, says its AI responses retrieve pages through its own “core Search ranking systems.” Two different underlying result sets would be expected to yield different citations, before any difference in models or freshness is considered.
The rest of the gap comes from how each system treats recency, whether it searches at all, and how each model chooses among the pages it finds.
Different crawlers and different indexes
Google AI Overviews are built on Google’s index. Google’s AI optimization guide says the system grounds responses by “relying on our core Search ranking systems to retrieve relevant, up-to-date web pages from our Search index.” Its AI features documentation says robots.txt rules for Googlebot are the control for crawling.
OpenAI documents its own crawler for ChatGPT’s search features. OpenAI’s crawler documentation says “OAI-SearchBot is used to surface websites in search results in ChatGPT’s search features,” and that sites opted out of it “will not be shown in ChatGPT search answers.” Seer’s finding suggests that the result sets ChatGPT draws citations from line up closely with Bing’s top results. The study’s sample was small, and OpenAI’s crawler page does not describe its search providers, so treat the Bing connection as a measured pattern rather than a documented architecture.
Two consequences follow. A page can be eligible for one system and blocked from the other, because the controls are separate: Googlebot for Google’s AI features, OAI-SearchBot for ChatGPT search. And a page ranking well in Google but poorly in Bing may show up in AI Overviews and not in ChatGPT, or the reverse.
Different appetite for recency
The two systems sit at opposite ends of a freshness study. Ahrefs compared the age of nearly 17 million cited URLs with organic results and found the four AI assistants it studied citing content 25.7% fresher on average, with ChatGPT showing the widest gap: its cited URLs were 393 to 458 days newer than organic results. Google’s AI Overviews were the exception, citing content that was 16 days older than organic results on average.
So for the same question, ChatGPT is, on average, more likely to cite a recently published or updated page, while the age of pages cited in AI Overviews tracks traditional Google results more closely. A page that is strong but old may appear in AI Overviews and not in ChatGPT for that reason alone.
Different decisions about when to search
ChatGPT does not search the web for every question. When it answers from what its model already knows, the sources it names, if any, come from training rather than a live search, and those can differ entirely from what any search engine would return today. The difference between knowledge a model learned in training and knowledge it retrieves at answer time is a large topic of its own. Here it matters only as one more reason the two systems diverge.
Google AI Overviews, for their part, do not appear on every search. Google says they are “only shown when our systems determine that it is additive to classic Search, and as such, often don’t trigger.” The two systems can disagree simply because one produced an AI answer for a query and the other did not.
Different models, different selection
Even with identical candidate pages, two different models would choose and present sources differently. Neither OpenAI’s nor Google’s documentation explains how its system weighs one retrieved page against another. Add normal variation within each system, which a SparkToro study found to be large for recommendation lists, and comparing a single ChatGPT answer with a single AI Overview tells you little.
What to do if you want to appear in both
| Factor | For Google AI Overviews | For ChatGPT search |
|---|---|---|
| Crawler access | Allow Googlebot | Allow OAI-SearchBot |
| Where to rank | Google organic results | Bing results matter, per Seer's February 2025 observation |
| Freshness | Cited pages' age close to classic results, per Ahrefs | Newer pages cited on average, per Ahrefs |
| Site-level opt-out | Search Console's Search generative AI control | robots.txt rule for OAI-SearchBot |
| Measurement | Search Console's Generative AI performance report | Repeated prompts; referral data in analytics |
1. Check both crawlers. Confirm robots.txt allows Googlebot and OAI-SearchBot on the sections you want cited. OpenAI says robots.txt changes take about 24 hours to apply.
- Watch Bing as well as Google. Use Bing Webmaster Tools alongside Search Console.
- Update important pages with genuinely new information, and show when they were updated.
- Measure separately. Google’s Generative AI performance report covers AI Overviews and AI Mode only. For ChatGPT, use repeated prompts and your analytics. Google Analytics 4 groups visits from sources such as ChatGPT into an AI Assistant channel, which by Google’s own description excludes AI Overviews and AI Mode.
Optimize the shared foundation first: useful, current, crawlable content. Then treat each system’s access, ranking environment and measurement as its own task, since the two start from different result sets, weigh recency differently, search under different conditions and run different models.