The tech giant Meta has begun collecting information about its employees' work, using data from their computers to accelerate the training and improvement of its neural networks.
New Data Collection Initiative
Meta Corporation has introduced an innovative tool called the Model Capability Initiative. This system is designed to collect data on employee actions performed on work computers. It covers various platforms, including Microsoft's GitHub, Salesforce's Slack, Atlassian, as well as Meta's own services — Threads and Manus.
A Meta spokesperson confirmed the launch of this project. He explained that real examples of PC usage are necessary for algorithms to create effective virtual assistants. Strict precautions were emphasized to protect confidential content, and the data is not used for any other purposes.
Ambitions of Mark Zuckerberg
The collection of this information is directly linked to Mark Zuckerberg's ambitious drive. He aims to close the gap between Meta and powerful competitors in the field of generative artificial intelligence, such as OpenAI, Anthropic, and Google. Earlier this month, Meta's Superintelligence Labs already presented its first major model called Muse Spark.
Assurances for Employees
In an internal memorandum, the company's management assured its employees regarding the operation of the new system. It was clearly stated that the system records only the screen content, without reading files or attachments. To avoid personal information being included in the database, employees are advised to refrain from solving private tasks on corporate devices.
What is Generative AI?
Generative artificial intelligence is a special type of AI systems capable of creating entirely new content. This can include text, images, audio, or even software code generated based on user queries. Such models are trained on vast datasets, allowing them to identify patterns and produce unique, meaningful results.
