Google's conversational AI chatbot Bard has recently been under the spotlight as SEO consultant Gagan Ghotra observed that Google Search had begun to index shared Bard conversational links into its search results pages. This development raised concerns as it could potentially expose information that users intended to be confidential. The conversations, even those relating to private emails, could be accessed and viewed publicly if users had shared the link with a third party.
Google's Response and User Reactions
Upon discovering this, Ghotra took to X, formerly known as Twitter, to share evidence of several Bard conversations being indexed by Google Search. The conversation gained traction, and Google Brain research scientist Peter J. Liu responded, clarifying that only those conversations that users had elected to share were being indexed. Google's Search Liaison account on X also addressed the issue, stating, “Bard allows people to share chats, if they choose. We also don't intend for these shared chats to be indexed by Google Search. We're working on blocking them from being indexed now.” Despite Google working on a fix, this incident has sparked a wave of concern among Bard users regarding privacy and data security.
https://twitter.com/gaganghotra_/status/1706545667362210129
Implications and Precautions
The incident has heightened the urgency of privacy concerns, especially since Bard is used for various applications, including customer service and personal assistance. Users are advised to exercise caution and avoid discussing sensitive topics and information that they wouldn't want to be tied back to them publicly. The public visibility of these conversations underscores the importance of being vigilant about the information shared with AI chatbots. Google has advised users that by refraining from using the share feature, they can ensure their chats don't end up in search results.
Just yesterday I reported that the Bard integration in Gmail is hallucinating and writing emails that do not exist. “Bard, in its current state, isn't quite the helpful inbox assistant that Google envisions,” noted a New York Times report, after the publication discovered these discrepancies during testing. The AI was found to create entire email conversations, raising concerns about its reliability and the potential implications for users.
