Meta Is Using Employee Computer Activity To Train Future AI Agents

Meta Platforms has decided to collect detailed records of Us employees' computer usage for training its AI models. The latest documents related to the same suggest that Meta is set to capture non-US data in the process as well, according to a report by Reuters. The documents introduce fresh issues for the project, a major component of Mark Zuckerberg's broader plan to transform how Meta operates around AI agents. In the weeks of its launch, Meta employees have complained that MCI was consuming a lot of data, and it was causing home internet usage to skyrocket. Some of the employees complained that it exhausted their month's data in a few days. Meta has also accepted in a question and answer document that the tool will capture the contents of any

emails or direct messages sent to United States personnel, regardless of the sender's location.Sam Altman's OpenAI Updates White House On New Biodefense AI Model Meta spokesperson Dave Arnold said MCI was installed only on US employees' devices like its focus on how people interact with computers, not the content of their screens. Arnold said, 'In the interest of transparency, we notified non-U.S. employees that it was deployed on the computers of U.S. colleagues they may email or chat with in the normal course of business.' He further said, 'We carefully considered and mitigated potential privacy risks in both the development and deployment of this tool, and we are committed to complying with applicable laws and regulations.'One employee from Meta also shared his findings in a detailed analysis of MCI log files performed with Claude. As per her, MCI was integrated on the company's existing data security software, providing it access to additional details like employees' code changes, sleep and wake cycle of computers, URLs visited, and content on the clipboard. Furthermore, this content is stored in low securely, unencrypted form. She wrote, 'Not ‘an AI that clicks a dropdown for you’ but ‘an AI that knows which dropdown to click, what to select, which document to paste it into, and what to do next.' The major concern of employees is that this data could be trained to pull jobs that they are needed for.