AI chatbots have become part of everyday life, with users sharing vast amounts of information with them while asking questions and discussing work and personal life. A new tool from Proton highlights the sheer volume of information these services can infer about their users.
Hot Topics
This Swiss company launched a free tool called "AI Paper Trail", which aims to show what artificial intelligence robots can know about the user.
The tool works by analyzing conversation logs that can be exported from OpenAI's ChatGBT and Anthropic's Cloud services.
To use the tool with GBT Chat, the user exports their data from the account settings, then obtains a compressed file containing their data, and then uploads it to AI Paper Trail for analysis.
Proton says that the data shared with the Lumo AI analytics platform is deleted after the process is complete and is not stored on its servers, and that the resulting report is available to the same user.
The importance of this tool lies in the fact that it not only displays the information that the user has explicitly written, but also reveals the picture that artificial intelligence can form about him through the accumulation of conversations.
According to the newspaper's experiment, ChatGBT may be able to know the user's name and occupation, understand how he uses artificial intelligence in his professional and personal life, as well as his geographical location, hobbies, relationships, and some of the devices he uses.
The experiment shows that frequent conversations over months allow the robot to create a very accurate profile of the user.
According to the newspaper, information that seems simple in a single conversation may become more significant when combined with dozens or hundreds of other pieces of information, and this shows that the user may reveal more about himself to artificial intelligence than he imagines.
The AI Paper Trail tool also offers tips for reducing risks, including sharing only the minimum necessary information, avoiding disclosing too many details about family and location, reviewing privacy settings, disabling chat history or using data to train models when the option is available, and periodically deleting chats and data.
Temporary chats can also be used when discussing topics that the user does not wish to keep a record of.
However, AI knowing so much about the user is not always a disadvantage, as the more the robot understands the user’s interests, context, and history, the more it is able to provide more personalized and accurate answers.
Therefore, there is a trade-off between privacy on the one hand, and having a smarter and more user-focused experience on the other.
Proton is using this initiative to promote its Lomo service, which it says focuses on privacy and does not keep or use chat logs to train AI models, and that there are other alternatives that focus on protecting user data.
The Proton tool reveals that the history of conversations with artificial intelligence can be transformed into a comprehensive and accurate picture of the user's personality and life.
Tags:
technology
