Shared conversations with Anthropic’s Claude chatbot were briefly accessible through Google search results because the webpages lacked noindex tags, according to The Decoder. This oversight allowed some private chats, including those containing crypto keys and legal questions, to be inadvertently exposed.
The Decoder also noted that OpenAI faced a similar issue last year, highlighting ongoing challenges in protecting AI chatbot data from unintended public indexing. Both incidents underscore the importance of robust data privacy controls in AI services.
For Japanese markets, where data privacy and security are critical amid growing interest in AI-driven financial tools, such leaks raise concerns about safeguarding sensitive information in FX, crypto, and equities platforms.