Alphabet’s Google, the largest customer of Scale AI, plans to sever its relationship with Scale AI following news that Meta has taken a 49% stake in the AI data-labeling startup. According to sources familiar with the matter, Google had intended to pay Scale AI approximately $200 million this year for human-labeled training data, which is essential for developing advanced AI models like Gemini, its competitor to ChatGPT. However, the search giant has already begun discussions with several of Scale AI’s competitors to shift much of its workload elsewhere. Scale AI’s significant loss of business coincides with Meta’s substantial investment, valuing the company at $29 billion, up from its previous valuation of $14 billion. Despite this, Scale AI intends to continue its operations while its CEO, Alexandr Wang, and a few employees transition to Meta. Given that its core business relies heavily on a few major customers, losing key clients like Google could severely impact Scale AI. In a statement, a Scale AI spokesperson emphasized the strength of its business, which includes partnerships with major companies and governments, while declining to comment specifically on its relationship with Google. Scale AI generated $870 million in revenue in 2024, with Google spending around $150 million on its services last year. Other major tech companies, including Microsoft and Elon Musk’s xAI, are also distancing themselves from Scale AI due to concerns about exposing proprietary data and research priorities to Meta, now a significant shareholder. This shift creates opportunities for Scale AI’s competitors, such as Labelbox and Handshake, to capture fleeing customers. Many AI labs are now considering hiring in-house data-labelers to ensure data security. Founded in 2016, Scale AI provides critical labeled data for developing sophisticated AI tools. The Meta deal benefits Scale AI’s investors, including Accel and Index Ventures, and positions its CEO, Wang, to lead Meta’s AI initiatives. This development marks a turning point for neutrality in AI development.
— new from CNBC
