AI assistants disagree on which brands they mention

If you ask four different AI assistants about the same product choice, you will not necessarily get the same brand names. That is shown by a new analysis from LLM Scout, published on October 7, 2026.

LLM Scout is an analytics platform that tracks how companies and brands are mentioned by AI systems. It is run by London-based AI consultant Frank Vitetta.

What was examined

The analysis is based on 749 identical questions about choosing between products. They were put to ChatGPT, Claude, Gemini and Perplexity. In total, it includes almost 27,000 AI answers that LLM Scout has recorded over nine months.

In 27.2 percent of cases, the four assistants disagreed on whether a particular tracked brand should be mentioned. That is, in more than one in four cases.

In other words, a brand can be highlighted by one assistant and not appear at all with another, even though the question is exactly the same.

In the study, ChatGPT was the assistant that most often mentioned a particular brand when the other models did not.

What it means

For ordinary users, the point is that an AI answer is not a neutral top list. If you use a chatbot to find a car, a bank or a phone, you may get a different selection if you ask another one. It can be a good idea to compare several sources.

For companies, the message is that it is not enough to optimize for one AI. The analysis indicates that the four systems retrieve and rank information differently. You therefore need to work on visibility across several assistants. Professionals call this Generative Engine Optimisation, that is, making your brand easy for AI to find and mention.

A caveat

The figures come from LLM Scout itself, which sells tools for measuring exactly this kind of visibility. It is therefore a company analysis and not an independent, peer-reviewed study. We have not been able to find details about how the questions were selected.