A recent news report revealed that conversations from Anthropic's Claude AI could be accessed through a Google dork, an advanced search query, after users had shared them via active links. Anthropic quickly fixed the exploit to disable this type of indexing. The incident raises questions about the privacy of AI conversations, which depends partly on user controls. For ChatGPT, users can chat without an account by logging out, though advanced features and higher thresholds are unavailable. Signed-in users can also enable "Temporary Chat" by clicking the circular icon in the upper right, which turns the prompt field black. The article covers steps to tighten privacy for four major AI chatbots.