[SAN FRANCISCO] Scale AI is firing hundreds of employees from the data label business after the meta platform invested US$14.3 billion in startups and hired a CEO.
The company will cut 200 full-time employees, roughly 14% of the global workforce, and offer retirements, scale spokesman Joe Osborne said Wednesday (July 16). Scales will also stop working with 500 global contractors, he said.
The move said it aims to “simplify our data business and help us move faster,” and added a scale plan for staff in other areas, including enterprise and government sales. In a memo sent to scale employees Wednesday, interim CEO Jason Droge said the layoff was the result of a business with data labels that brought too many people last year. It led to “too many layers, excessive bureaucracy, useless confusion about the mission of the team,” Droge wrote in a note seen by Bloomberg News.
Droege added that “changes in market demand” also contributed to the decision to restructure. Following its agreement with Meta, some of Scale's most prominent clients have phased out work with startups, including Openai and Alphabet's Google, according to reports by Bloomberg and others.
Founded in 2016, Scale was the most well-known name on the market to help label and annotate the data needed to build artificial intelligence (AI) models. It generated approximately US$870 million in revenue in 2024, and is projected to earn US$2 billion this year, Bloomberg News reported in April.
In June, Meta completed an investment in the billion-dollar scale and acquired a 49% stake in the company. As part of the deal, co-founder Alexandr Wang left the startup to lead Meta's new Superintelligence Unit, part of the Facebook parent's multi-billion dollar investment to keep up with AI development.
BT in your inbox
Deliver the latest news stories and analysis directly to your inbox, starting and ending daily.
Despite its position as a market leader in providing the key ingredients needed to build AI models, Scale faces growing rivals, including Turing, invisible technology, Labelbox and Uber technology. Some Scale's clients said that competitive services are surged from customers as they were worried that Meta would increase additional visibility into the AI development process.
In a note, Droege said he hopes that the change will expand over the long term, become more efficient, allowing the part that avoids the data in the startup to “focus on the largest and most profitable opportunities.” He also said the new structure “can serve the customers we have today and bring back the customers who have slowed down working with us.”
Scale is one of several AI companies that saw key talent hired last year without acquiring it last year. Recently, Google signed a US$2.4 billion deal with AI coding startup Windsurf, employing a CEO and some of its top staff. The transaction raises the question of what happens to employees left after the CEO joins a large company.
In the case of Windsurf, the startup was quickly purchased by another AI company. On a massive scale, the company plans to capitalize on the hearts of war, Doroge said.
“We are still a rich and funded company, and today's announcement will help us accelerate new investments and add resources if necessary,” he wrote in the note. Droege said the startup plans to hone the scope of the business that learns data and focus on projects related to coding, language and speech.
Osborne said the company plans to hire hundreds of people later this year for efforts such as creating custom AI applications and working with US government agencies such as the Department of Defense. Currently, these parts of the business are earning nine figures in revenue, Droege previously told Bloomberg. Bloomberg
