AI Company Crisis Reveals Risks of User Data Leakage

Users often share their deepest secrets and personal concerns with chatbot AI, trusting that their information remains confidential. However, with an American AI company nearing bankruptcy, user data and private conversations are at risk of being sold, raising serious concerns about the fate of sensitive information. The crisis highlights that the collapse of AI companies could turn billions of people's secrets into commercial assets traded on the market, given the high costs of developing and operating such technologies. If these companies face bankruptcy or mergers, their user databases might be sold off as part of asset liquidation processes. The director of an AI and human rights program explains that users tend to disclose intimate personal details during conversations with human-like chatbots, increasing the volume of exposed information. Users may also share details about others, making data about non-users part of these databases. Court documents reveal that the company can offer user data to potential buyers, despite selling some assets to a major corporation excluding user data. A coalition of state attorneys general in the U.S. is pushing for safeguards to protect children's and users' privacy, but the lack of strong legislation limits these efforts. Privacy policies include exceptions allowing companies to relinquish certain commitments in bankruptcy, creating a loophole for data sales. Even regions with strict laws, like the European Union, do not necessarily prohibit data sales but require buyers to respect original privacy policies. The company uses personal data to train models, develop algorithms, create user profiles, and target advertisements. Data sensitivity increases when involving children and teenagers, especially with reports linking platform conversations to incitement of violence and suicide. In tragic incidents, a chatbot was accused of encouraging a teenager to commit violence, and a 14-year-old boy died by suicide after interacting with the platform. Experts advise reviewing apps that store sensitive information and deleting data from unused platforms, following privacy settings or contacting customer support. The crisis reveals a deeper truth: secrets entrusted to AI chatbots may not remain protected indefinitely, particularly if the owning company becomes insolvent.