Alexandr Wang, CEO of Scale AI, made headlines during a recent appearance on CNBC’s “Squawk Box” at the World Economic Forum in Davos, Switzerland, on January 23, 2025. This follows reports that Google, Scale AI’s largest customer, plans to sever ties with the AI data-labeling startup after rival Meta acquired a 49% stake in the company, a source revealed to Reuters.
Initially, Google had expected to allocate around $200 million to Scale AI this year for essential human-labeled training data crucial in developing advanced technologies, including its Gemini AI model, which competes with ChatGPT. Following the recent developments, Google has engaged in discussions with several of Scale AI’s competitors to transition much of that workload.
The impending loss of this significant business for Scale AI comes after Meta’s investment increased its valuation from $14 billion to an impressive $29 billion. In the immediate aftermath of the acquisition, Scale AI is determined to continue its operations, even as Wang and select employees move to Meta. However, experts caution that the company’s reliance on a narrow customer base could jeopardize its stability if it loses major clients like Google.
Despite the challenges, a spokesperson from Scale AI asserted that the company remains robust, emphasizing its commitment to data security across its partnerships with various corporations and governments. Specific comments regarding Google were not provided.
In 2024, Scale AI generated $870 million in revenue, with Google accounting for approximately $150 million of that total. Yet other tech giants, including Microsoft, have also begun distancing themselves from Scale AI. Reports indicate that Elon Musk’s xAI is exploring a similar exit, while OpenAI, which previously reduced its financial ties, stated it would continue collaborating with Scale as one of several data vendors.
Competitors express concerns that collaborations with Scale could inadvertently reveal proprietary data to Meta, thus compromising their research innovations. Many AI companies typically share confidential data and prototypes to receive Scale’s data-labeling services, raising alarms now that Meta is a significant stakeholder.
With Google having sought to diversify its data service providers for over a year, Meta’s recent acquisition has intensified these efforts. Sources indicated that Google may withdraw from all primary contracts with Scale AI swiftly due to the structure of these data-labeling agreements, creating opportunities for rivals eager to step in.
“The Meta-Scale deal marks a turning point,” stated Jonathan Siddharth, CEO of Turing, a competing firm. He noted that AI labs are increasingly recognizing the importance of maintaining neutrality when partnering with data providers. Manu Sharma, CEO of Labelbox, predicted that his company could earn hundreds of millions in new revenue from clients leaving Scale, while Garrett Lord from Handshake reported a surge in demand for his company’s services from AI labs competing with Meta.
In light of these developments, many AI companies are opting to hire in-house data-labelers, aimed at keeping their proprietary data secure. Brendan Foody, CEO of Mercor, emphasized that his startup not only competes directly with Scale AI but also creates technology to help recruit and evaluate candidates, allowing AI labs to quickly scale their data-labeling efforts.
Founded in 2016, Scale AI has played a pivotal role in providing labeled data essential for sophisticated tools like OpenAI’s ChatGPT. The Meta deal not only promises to benefit Scale AI’s investors, such as Accel and Index Ventures, but also positions Wang to lead Meta’s AI efforts as the company seeks to regain momentum after falling short in earlier AI advancements.
Original Source: https://www.cnbc.com/2025/06/14/google-scale-ais-largest-customer-plans-split-after-meta-deal.html
Category :
Tags:
Publish Date: 2025-06-15 03:13:00
