South Lake Tahoe, California: how many business websites block AI.
Of 37 reachable business websites in South Lake Tahoe in the census, 10.8% disallow at least one named major AI crawler in robots.txt.
Measured separately, by a different test: 16.2% (6 sites) blocked a request that claimed to be GPTBot. That request came from AIOInsights servers, not OpenAI's network, so a wall that checks crawler IP addresses can block it while letting the real GPTBot in. It is not added to the figures above.
Is your website on the wrong side of this? Run the free AIOInsights check: about 20 seconds, no account, and it says in plain words whether AI crawlers can read your site.
How the census was taken
AIOInsights took every business in California that lists a website in OpenStreetMap (map data as of 2026-11-01; contributors credited under the Open Database License), kept one record per website, and checked each site between 2026-11-01 and the publish date. For each site it read /robots.txt with the same rules the free AIOInsights check uses, and loaded the homepage twice: once as a browser and once identifying as GPTBot, OpenAI's crawler. A site counts as reachable when the browser visit loads; only reachable sites are in the percentages.
- Disallows a named AI crawler (the headline): robots.txt names a major AI crawler (GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, PerplexityBot, Google-Extended and their kin) and disallows the whole site. This is the site's own published instruction, read directly.
- Blocks AI search: robots.txt shuts out the crawlers AI assistants use to answer live questions (OAI-SearchBot, ChatGPT-User, Claude-SearchBot, PerplexityBot), or every crawler.
- Blocks every crawler:
User-agent: *withDisallow: /, which also shuts out Google. - Blocked a request that claimed to be GPTBot: reported on its own, never added to the headline. The homepage request carried GPTBot's user-agent but came from AIOInsights servers, not OpenAI's network. The site answered it with a bot challenge or a 401/403 while letting the browser visit in. Bot walls that verify crawler IP addresses turn that request away and still let the real GPTBot in, so this counts sites that refuse an unverified GPTBot request, not sites that block OpenAI.
This census publishes totals only. It never names a business. Any business can check its own site free, with no email, at aioinsights.com.