Google Plans to End Partnership with Scale AI Following Meta’s Major Investment, Sources Reveal

In a surprising turn of events, Google, the biggest customer of the artificial intelligence (AI) data company Scale AI, is preparing to cut ties with the startup. This decision comes after Meta, the parent company of Facebook, announced it was buying a 49% stake in Scale AI, according to sources familiar with the matter. The deal has raised concerns among other tech giants, including Microsoft and Elon Musk’s xAI, who are also reconsidering their partnerships with Scale AI.

Scale AI is a company that provides labeled data, which is essential for training AI models. These models power advanced technologies like chatbots, image generators, and self-driving cars. Google had planned to spend around $200 million this year on Scale AI’s services to improve its own AI systems, such as Gemini, its competitor to OpenAI’s ChatGPT. However, with Meta now owning nearly half of Scale AI, Google and other companies worry that their confidential AI research and strategies could be exposed to a major competitor.

Meta’s investment has increased Scale AI’s value to $29 billion, double its previous worth of $14 billion. Despite this financial boost, the company could face serious challenges if it loses key clients like Google. Scale AI’s CEO, Alexandr Wang, and some employees are expected to move to Meta, but the company insists its business remains strong. A spokesperson for Scale AI stated that they work with many large corporations and governments and are committed to protecting customer data. However, they did not comment specifically on Google’s plans to leave.

image

Last year, Google spent about $150 million on Scale AI’s services, contributing to the startup’s total revenue of $870 million in 2024. Other major tech firms, including Microsoft and xAI, are also stepping back from Scale AI. OpenAI, the creator of ChatGPT, had already reduced its reliance on Scale AI months ago, though it still uses the company as one of its many data providers. The chief financial officer of OpenAI mentioned that they would continue working with Scale AI but did not provide further details.

The main issue for these companies is the fear that Meta, as a major shareholder, could gain access to their AI development secrets. When businesses hire Scale AI to label data, they often share sensitive information about their projects. With Meta now heavily involved in Scale AI, competitors worry that their research plans and technical details might no longer be safe.

Scale AI earns most of its money by providing highly skilled human workers who label complex data for AI training. These workers include experts like historians, scientists, and even people with PhDs. Their job is to review and categorize information so that AI models can learn from accurate examples. As AI technology advances, the demand for such specialized data has grown, with some annotations costing as much as $100 each.

Apart from working with AI companies, Scale AI also handles data labeling for self-driving car firms and the U.S. government. These clients are likely to stay, but the biggest profits come from partnerships with generative AI developers. Google had already been looking to spread its data-labeling work across multiple providers over the past year. However, Meta’s recent investment has sped up Google’s decision to move all its important contracts away from Scale AI.

The shift could happen quickly because of how data-labeling agreements are structured. If Google and other tech giants follow through, Scale AI may need to find new customers to replace the lost business. Meanwhile, rival data-labeling companies see this as an opportunity to attract big clients looking for neutral providers.

The situation highlights the growing competition in the AI industry, where companies are racing to develop the most advanced models while keeping their strategies confidential. With Meta’s deep involvement in Scale AI, other tech firms are now reconsidering their reliance on external data providers. Some may choose to hire their own in-house teams to label data, ensuring that their AI research remains private.

As the AI industry continues to expand, partnerships and rivalries between major players will shape the future of technology. Google’s potential exit from Scale AI marks a significant moment in this ongoing battle, showing how quickly alliances can change in the fast-moving world of artificial intelligence.

image

Meta’s $14.8 Billion AI Partnership Faces US Regulatory Scrutiny

image

Taiwan Restricts US-Listed Chinese Tech Giants Huawei and SMIC Over Export Concerns