Four AI engines, four different ways of choosing brands.
ChatGPT, Perplexity, Claude and Google AI Overviews do not pick brands the same way: two answer mostly from what their models learned, one searches the web for every answer, and one builds on Google's search index. Proofsource asks all four the same buyer questions and tracks who each one names.
2 free scans, today and tomorrow: 25 questions on 4 engines. No card.
The engines Proofsource tracks on every plan
ChatGPT
Perplexity
Claude
Google AI OverviewsHow do the AI engines differ?
Where each engine gets its answers, whether it shows its sources, and which crawlers decide whether it can read your site. Each row links to that engine's page, with the vendor's documentation.
| Engine | Answers from | Shows sources | Crawlers that matter |
|---|---|---|---|
| ChatGPT | Model knowledge, plus ChatGPT search when it searches the web | When it searches, with links to the pages it used | OAI-SearchBot, GPTBot, ChatGPT-User |
| Perplexity | A live web search for every answer | Always, as numbered citations | PerplexityBot, Perplexity-User |
| Claude | Model knowledge, plus web search when it is used | When it searches, with links to its sources | ClaudeBot, Claude-SearchBot, Claude-User |
| Google AI Overviews | Google's search index, using query fan-out | Always, as links beside the summary | Googlebot, Google-Extended |
Pick an engine to see how it chooses
ChatGPTHow ChatGPT decides which brands to name
Without a search, ChatGPT answers from what its model learned in training. Brands that appear often on pages it learned from are the ones it reaches for.
PerplexityHow Perplexity picks the brands it names
Perplexity runs a search for each question, reads the top pages, and writes an answer with numbered citations.
ClaudeHow Claude decides which brands to recommend
By default Claude answers from its training. A brand with little written about it on pages Claude could learn from is easy for it to miss or mistake.
Google AI OverviewsHow Google AI Overviews choose the brands they name
Google says a page must be indexed and eligible to show with a snippet to appear in AI Overviews or AI Mode, and that no special files or markup are needed.
Why track every engine, not just ChatGPT?
Buyers do not all use one assistant, and the engines disagree. A brand can be a regular in Perplexity's cited answers and missing from Claude's, because one reads the web live and the other answers from training.
Asking all four the same questions shows where the gap is engine-specific (a crawler blocked, a page not indexed) and where it is everywhere (the pages your category relies on do not mention you). The fix is different in each case.
The terms used here are defined in the AI visibility glossary, and how each number is measured is on the methodology page.
What people ask about AI engines and brands
Which AI engines does Proofsource track?
ChatGPT, Perplexity, Claude and Google AI Overviews, on every plan, with the same questions on each. Grok and DeepSeek are available as paid add-ons. Gemini, Microsoft Copilot and Google AI Mode are not tracked today.
Do the AI engines name the same brands?
Often the same leaders, but not in the same order or as often. In our own scan of one category, the most-named tool appeared in 10 to 13 of 20 answers depending on the engine, and only one of the four engines described a brand-new company correctly.
Should I block AI crawlers?
Blocking a search or user fetcher (OAI-SearchBot, PerplexityBot, Claude-SearchBot, Claude-User) keeps your pages out of those engines' answers. Training crawlers (GPTBot, ClaudeBot, Google-Extended) are a separate choice about what future models may learn. Check your CDN too: a bot-blocking setting there can override robots.txt.
How often does Proofsource ask each engine?
Every day on paid plans, once per question per engine. The free trial runs 2 scans, today and tomorrow, with 25 questions on each engine.
- OpenAI: overview of OpenAI crawlers developers.openai.com/api/docs/bots
- Perplexity: crawlers docs.perplexity.ai/guides/bots
- Anthropic: does Anthropic crawl data from the web support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler
- Google Search Central: AI features and your website developers.google.com/search/docs/appearance/ai-features
- Google: common crawlers (Google-Extended) developers.google.com/search/docs/crawling-indexing/google-common-crawlers
Sources read on .
See what all four engines say about you.
See which buyer questions name your brand, who gets named instead, and what to fix. 2 free scans, today and tomorrow: 25 questions on 4 engines. No card.
Last updated