DuckDuckGo has published fresh survey data revealing that many users routinely divulge deeply personal information to AI chatbots—even things they’d never share with friends, family, or doctors. Among nearly 2,000 adult respondents in the U.S., 31% admitted they’ve disclosed such sensitive details to an AI, and the percentage jumps to 56% among self-described AI enthusiasts. Some users aren’t even aware how much of their interaction with these tools is exposed.
What Users Don’t Know
The survey found that over half of participants are unaware or uncertain that most AI chatbots use conversations for ongoing training by default; only a quarter understood that their AI chats might be legally accessible through subpoenas. When informed about how their data are being used, 58% reported discomfort; 30% said they were very uncomfortable with how their conversations might feed into AI training.
Parents vs. Non-Parents — A Surprising Gap
There’s a notable divide between parents and non-parents in how openly they communicate with AI systems. Parents or guardians with children in the home are almost twice as likely to tell an AI something they wouldn’t share with another person—43% did so, vs. just 25% among users without children at home.
On trust, respondents generally placed slightly more faith in large AI firms than in the U.S. government when it comes to safeguarding their data. But skepticism remains strong: 39% said they have no trust at all in big AI companies, while 48% expressed complete distrust of the government’s role in protecting their conversations.
DuckDuckGo emphasized there’s still room for change. Although many feel their information is already out there, the company argues that protecting privacy begins with each new conversation. Users are encouraged to use AI tools that clearly explain how data are handled and to be mindful whenever they share personal data with chatbots.
As AI-chat habits become more deeply ingrained in daily life—whether for work, education, or companionship—the implications of sharing personal information become all the more important. This survey underscores how much more transparent AI providers need to be. Users deserve clear controls and safeguards. The next frontier in AI privacy may depend on new policies, user education, and possibly regulation to ensure conversations meant to be private truly stay private.
