Meta plans to use data from its employees' mouse movements and keystrokes to train its AI models, according to a Reuters report. The company will collect this information through an internal tool that captures inputs on certain applications. This move follows a trend in the tech industry where companies are seeking new sources of training data for their AI models.
The data collected by Meta will not be used for any purpose other than training its AI models, and safeguards have been put in place to protect sensitive content, according to a company spokesperson. The spokesperson stated that using real-life examples of how people interact with computers is necessary to help train the models effectively.
Meta's move raises concerns about employee privacy as companies increasingly rely on user data to develop their AI models. Last week, it was reported that old startups were being scoured for corporate communications such as Slack archives and Jira tickets, which are then converted into AI training data.
The use of internal employee data by tech companies highlights the growing reliance on user data in the development of artificial intelligence. This trend has significant implications for privacy laws and regulations as the industry continues to evolve.
