Anthropic's Claude AI experienced a privacy incident where user-generated session sharing links were publicly indexed by search engines like Google and Bing, allowing anyone to view conversation content through searches. This was a technical configuration oversight, not a backend database breach, as Anthropic failed to effectively prevent search engine crawlers from accessing the sharing links. The /share/ path of Claude was set to be disallowed from crawling in robots.txt, but the page response header also set the X-Robots-Tag: none attribute, creating a contradiction where Google's documentation states that if a page is blocked by robots.txt, crawlers cannot read the noindex tag, leading to URLs being discovered via external links and included in search results. Tested indexed sessions contained sensitive information such as wallet private keys, illegal consultation records, internal company tasks, and content identifying personal identity. Anthropic has communicated with major search engines to remove links and revised its strategy to completely prohibit crawling, though some cached or residual links remain.