Can I block AI crawlers and still rank in Google?
Short answer: Yes. Blocking GPTBot, ClaudeBot or PerplexityBot has no effect on Google rankings, but you cannot opt out of AI Overviews without also leaving Google Search.
Yes. Google's rankings depend on Googlebot, and the AI crawlers run by OpenAI, Anthropic and Perplexity are entirely separate. You can disallow GPTBot, ClaudeBot and PerplexityBot in robots.txt and your Google rankings will not change.
There are two important caveats.
You cannot block AI Overviews separately. Google AI Overviews and AI Mode are built from the same index as Search. The Google-Extended token only controls whether your content trains Gemini models; disallowing it does nothing to AI Overviews. The only way out of AI Overviews is noindex, which also removes you from Search.
Blocking AI crawlers has a cost of its own. If ChatGPT, Claude or Perplexity cannot read your pages, they do not fall silent about you. They answer from older training data or from third-party sites (directories, review platforms, competitors) that describe you, often inaccurately. For most service businesses and brands that is worse than being read.
A middle path many organisations choose is to separate training from retrieval:
# Opt out of training, stay citable
User-agent: GPTBot
User-agent: ClaudeBot
User-agent: Google-Extended
Disallow: /
User-agent: OAI-SearchBot
User-agent: Claude-SearchBot
User-agent: PerplexityBot
Allow: /
This keeps your content out of future model training while still allowing the search-oriented bots to fetch and cite your pages when a user asks.
Whatever you decide, make it deliberate. In audits the most common finding is a site that blocks every AI bot by accident through a wildcard rule added years ago. Seoptist's audit lists exactly which user agents are allowed or blocked and flags firewall-level blocks that robots.txt alone would not reveal.