Google’s AI Overview produced the broadest coverage, with more than six brand mentions on average for each query. ChatGPT’s answers tended to be shorter, closer to two mentions, while AI Mode dropped the number even lower. The gap matters most on searches that suggest buying intent. Terms with “buy,” “where,” or “deals” in them pulled brand names about two-thirds of the time across all three tools, echoing the way e-commerce has always relied on keyword intent to surface sellers.
The pattern isn’t random. ChatGPT leans on names that show up frequently in its training data, which favors brands with long exposure in public information. Google’s Overview tool casts a wider net, including more companies in each response. AI Mode trims the list instead, putting more weight on a smaller group of brands it deems worth citing. Industries behave differently too: healthcare and education queries had some of the widest disagreements, while finance and online retail were a little closer, but still far from consistent.
There’s also a quirk in how the platforms handle citations. Google tends to show its sources more often than the number of brands it actually names, building a visible trail back to original material. ChatGPT works the other way, mentioning brands more than it cites, which suggests it is relying on recognition baked into its training data rather than live source pulls. That makes visibility uneven, companies might get surfaced without a traceable citation, which doesn’t happen as much on Google’s tools.
For companies hoping to show up in these answers, the study suggests no single path works across the board. Traditional search optimization is still the base strategy, since ranking in established search feeds the material that Google’s AI draws from. At the same time, brand awareness itself plays a larger role than before. Longstanding names appear in ChatGPT even without fresh signals, while Google’s systems open the door for more variety if a brand can find its way into citation networks.
What emerges is not one uniform system but three very different filters. Sometimes they overlap, but most of the time they don’t, which means visibility in AI-driven search is already fragmenting. The way brands adapt to that split may decide whether they gain ground in this new discovery layer or watch competitors get there first.
Read next: Amnesty Warns Big Tech’s Power Now Threatens Privacy, Rights, and Competition Worldwide