
Image credit: Search Engine Journal
ChatGPT frequently generates specific brand and product names in its initial search queries, even when users do not mention them, fundamentally altering strategies for search engine optimization and artificial intelligence visibility, according to new research.
This pre-selection process, observable through browser developer tools, suggests that inclusion in ChatGPT’s initial shortlist is significantly more valuable than general web findability, estimated to be 33 times more impactful.
The generative AI model, developed by OpenAI, crafts its own search queries that users can view, often containing brand and product names not prompted by the user’s initial request, researchers reported.
ChatGPT then conducts what are described as ‘fan-out’ searches, performing individual site-specific queries for each brand on its predetermined shortlist.
This behavior means that brands like Roborock, Saros, and Dreame X50 could be specifically targeted in AI-generated searches for robot vacuums, even if a user’s query was generic.
The model’s knowledge extends to current product models and can differentiate between searching for vendors versus reviewers based on the product category, according to the research.
For instance, when asked about accounting software, ChatGPT’s internal queries might include specific brands such as Xero, QuickBooks, Zoho Books, FreshBooks, Wave, and Sage.
Similarly, for language learning applications, the AI could generate queries for Duolingo, Babbel, Busuu, Pimsleur, Speak, and LingQ.
This mechanism also applies to web hosting services, potentially shortlisting Hostinger, SiteGround, Cloudways, and Kinsta, or therapy platforms like BetterHelp, Talkspace, and Brightside Online-Therapy.com.
The findings indicate a shift in how companies need to approach digital marketing, moving beyond traditional SEO to focus on AI visibility and direct inclusion in these pre-selection algorithms.
Organizations mentioned in the research, including Granola, Notion AI, Otter, Fireflies, Fathom, Mem, Limitless, Intercom, Fin, Zendesk AI, Ada, Decagon, and Sierra, represent a range of tools and services that could be subject to this AI-driven pre-selection.
The study highlights that Google, a major player in search, is also influenced by these AI-driven query behaviors.
Source: Search Engine Journal
Written by
Joyce de Castro
Joyce is a core team member at Rabbit Rank and the lead author covering SEO news, algorithm updates, industry trends, and actionable ranking strategies.
Keep reading
Related Articles

OpenAI says robots.txt may not apply to ChatGPT’s fetch bot
OpenAI argues robots.txt rules may not apply to ChatGPT’s fetch bot for user-initiated requests, leading to wi...

Anthropic to Watermark Claude AI Text for EU Users Under New Act
Anthropic plans Claude AI watermarking for EU AI Act compliance, sparking debate among writers and developers...

YouTube Promotes Creator Search Lift Amid Attribution Challenges
YouTube is promoting creator partnerships based on branded search lift, but a new report reveals significant a...