Scale AI CEO: Turning AI Investments Into Real Business Value

Despite widespread enthusiasm for artificial intelligence, most companies are not yet seeing meaningful returns. A recent MIT report found that 95% of businesses adopting AI fail to generate revenue from their projects, fueling concerns that the industry could be a bubble. Jason Droege, the new CEO of Scale AI, believes better execution and problem alignment can change that trajectory. 

Scale AI, best known for its data-labeling services that train large language models (LLMs), has become a pivotal player in the ecosystem. In June, Meta purchased a 49% stake in the company for $14.3 billion, valuing Scale at $29 billion. The deal triggered leadership changes, with founder Alexandr Wang stepping aside and Droege, a former Uber executive, taking over. While some rivals like OpenAI and Google reportedly pulled back from partnerships after Meta’s investment, Scale says its data business continues to grow month by month. 

Droege is now steering Scale toward helping enterprises create custom datasets and AI-driven applications tailored to business needs. Instead of chasing generalized chatbot hype, Scale is working with organizations including the Mayo Clinic, Cisco, the Qatari government, and the U.S. Defense Department — the latter via a $99 million Army contract — to automate document-heavy processes and streamline decision-making. 

Key takeaways for enterprises include: 

• AI success depends on fit: Companies gain value when AI addresses areas where humans are slow, inconsistent, or error-prone. 

• Human expertise remains essential: Senior professionals must refine and validate AI outputs to ensure accuracy and trust. 

• Returns take time: Analysts predict it could take years before AI generates broad organizational revenue. 

While competition from Amazon, Microsoft, and thousands of startups looms, Droege argues that Scale’s dual strength in data and applications positions it to help businesses finally unlock AI’s long-promised value. 

 

Source: 

https://edition.cnn.com/2025/09/30/tech/scale-ai-making-money-meta