A recent report details serious concerns about Grok, Elon Musk’s AI chatbot, which has made over 370,000 conversations publicly indexed on various search engines including Google, Bing, and DuckDuckGo. This issue reportedly exposes various sensitive topics, including discussions on drug manufacturing and even a plan for the assassination of Elon Musk.
Grok users create shareable links that inadvertently become indexed online, with no prior notification to the users that their conversations could be made public. Notably, Forbes noted that many user chats have included illegal topics such as drug production methods, malware coding, and instructions for making bombs. A search for these contents reveals numerous Grok interactions that should not be publicly accessible.
Despite provisions in Grok’s service agreement that prohibit the promotion of violence or harm, users have uncovered that these safeguards either malfunctioned or were absent, raising significant fears about the efficiency of Grok’s moderation controls.
Related Articles:
- Elon Musk claims Grok was ‘manipulated’ into praising Hitler
- ChatGPT removes display conversations feature due to indexing concerns
- Musk’s Grok response fails to meet safety expectations
As investigations continue, concerns grow over the implications of allowing AI technologies to operate without stringent oversight, especially those as influential as Grok.
