Meta eyes $10 billion investment in Scale AI to boost model training capabilities

Founded in 2016, Scale AI specialises in data labelling, a critical process that involves accurately annotating and classifying data to improve the performance and efficiency of AI systems.

author-image
Social Samosa
New Update
Meta Eyes $10 Billion Investment in Scale AI to Boost Model Training Capabilities

Meta is reportedly in advanced discussions to invest as much as $10 billion in artificial intelligence startup Scale AI, as part of its broader efforts to expand its AI capabilities, according to a report by Bloomberg.

Founded in 2016, Scale AI specialises in data labelling, a critical process that involves accurately annotating and classifying data to improve the performance and efficiency of AI systems. The company currently collaborates with several leading tech firms, including Meta and OpenAI, in supporting AI model development.

Meta, which previously contributed $1 billion to Scale AI’s Series F funding round, is said to be seeking a deeper partnership with the company. The potential deal could offer Meta exclusive access to Scale’s latest data classification tools, resources that could substantially enhance the performance and accuracy of Meta’s AI models.

The move aligns with Meta’s wider infrastructure expansion. The company is building a 2-gigawatt data centre and currently possesses an estimated 350,000 Nvidia H100 chips, surpassing holdings by both OpenAI and Elon Musk’s xAI, which reportedly have around 300,000 chips each. Meta is also in the process of developing proprietary AI hardware.

Industry analysts suggest the investment signals Meta’s ambition to become a dominant force in the global AI race. A stronger AI backbone could affect a range of Meta’s platforms and products, including Facebook, Instagram, its virtual reality initiatives, and future wearable technologies.

 

Data labelling Model training Scale AI