Meta company is using new software called Model Capability Initiative to monitor employees' keyboard strokes, mouse usage, and even computer screen images in order to train their AI models to become more advanced.
The initiative is mandatory to install on work laptops of employees and contractors in the United States, with absolutely no opt-out option, as confirmed by Meta's Chief Technology Officer Andrew Bosworth.
This monitoring will include work-related websites and apps like Gmail and GChat, and the AI will learn in detail how people use computers and how they use shortcuts.
According to the company, this data will only be used for training AI and will not be used in assessing employees' work performance. However, among employees, there is considerable dissatisfaction and concern about this excessive monitoring.
Experts in technology and digital rights are criticizing this as the worst workplace privacy violation.
"Employees are essentially training the AI systems that will replace their jobs," some experts have pointed out. This could also mean that Meta is preparing for AI to completely replace human functions in the future.
On the other hand, Meta is investing over $135 billion in AI technology, and starting from May 20, is planning to lay off approximately 8,000 employees (10 percent of the total workforce).
Since 2022, Meta has reduced over 25,000 jobs, so criticism is intensifying over the practice of recycling employees' work functions as data for AI.
Ref: Cnet
