Key Takeaways:
- Wang stepped down as CEO of Scale AI to lead Meta’s new superintelligence initiative following a $14.3 billion investment.
- He believes AI’s next leap won’t come from a single breakthrough, but through steady advances in data, infrastructure, and agent design.
- Wang envisions a near future in which AI agents drive most economic activity, with humans supervising their output.
At 28, Alexandr Wang is entering a new phase of his career—not by founding another company, but by joining one of the largest. After nearly a decade leading Scale AI, the data infrastructure company he co-founded at age 19, Wang is now heading up Meta’s newly launched superintelligence team.
His appointment follows Meta’s $14.3 billion minority investment in Scale AI, a move that not only signaled the company’s ongoing commitment to data but also solidified Wang’s role in shaping AI at the highest levels.
Wang, who once held the title of youngest self-made billionaire, told TIME his decision came down to impact. “I started Scale to accelerate AI,” he said. “But I believe we’re entering a new phase—where the bottleneck is no longer just data, but what we build with it.”
Data-First Discipline
Wang has long viewed data as the backbone of artificial intelligence. That insight shaped Scale AI into a critical infrastructure provider for companies such as Meta, Microsoft, OpenAI, and major automakers and defense agencies. From image annotation to model evaluation, the company has built systems to support nearly every part of the model development lifecycle.
But Scale’s aggressive scaling wasn’t without criticism. The company relied on a vast network of contract workers—reportedly over 240,000—for tasks like data labeling and model scoring. Some former employees, including cofounder Lucy Guo, raised concerns about delayed contractor payments. Wang acknowledged the tension but defended Scale’s approach, suggesting that operational challenges were addressed as the company matured.
“There’s always a trade-off when you grow fast,” Wang said, “but our core values never changed—we were obsessed with quality, speed, and building things that matter.”
Vision: An Agentic Economy
Wang now sees the next step in AI’s evolution not just in better models, but in the emergence of AI agents—software entities capable of completing tasks autonomously. These agents, he believes, will eventually handle a majority of the world’s economic activity.
He refers to this vision as the “agentic world,” where agents take over execution and humans shift into higher-level oversight. “It’s not just about automation,” he explained. “It’s about delegation. AI will manage supply chains, customer service, research, scheduling—entire workflows—with humans reviewing, steering, and refining.”
But realizing that future, Wang warns, requires infrastructure that doesn’t yet exist—systems for decision review, trust, resource allocation, and more. “We need to build the rails before we let the train run,” he said.
Superintelligence: A Long-Term Project
Unlike many in Silicon Valley, Wang does not believe artificial general intelligence (AGI) is just around the corner. Instead, he compares the path to curing cancer—complex, nonlinear, and filled with narrow victories.
“There won’t be one AGI model that does everything,” he said. “Different systems will specialize, and each will need its own data flywheel, evaluation protocols, and deployment standards.”
Wang described the current state of AI as “brittle” when stretched across domains. A model that excels at summarizing documents might struggle to reason over video or audio. That’s why he believes the focus must remain on reliability, tooling, and building capabilities one layer at a time.
“It’s not the moment for shortcuts,” he said. “It’s the moment for system thinking.”
Leadership Style and Cultural Discipline
During his time at Scale AI, Wang earned a reputation for running a tight operation—known for intensity, precision, and long hours. But he also emphasized the value of clarity and culture.
“You can’t achieve big things without a mission people believe in,” he said. “Our goal wasn’t just to grow. It was to build a foundation for the future of intelligence.”
As for his new role at Meta, Wang brings a similar mindset. He’ll oversee a team focused not only on pushing model performance, but on building the software, interfaces, and alignment layers required for true superintelligent systems.
What Comes Next
Wang’s move to Meta marks a rare transition: from founder-CEO to executive builder inside a tech giant. It also reflects growing recognition that AI development is shifting from research breakthroughs to operational excellence.
Whereas the past decade prioritized open-source models and novel architectures, the next may hinge on execution—how well organizations integrate agents, monitor decisions, and align outcomes with real-world constraints.
Wang is betting that superintelligence won’t be a lightning strike. It will be the result of slow, careful construction, piece by piece. And whether that’s within Meta or across a broader ecosystem, he seems content to build—rather than chase hype.
“We’re just getting started,” he said.
Learn how AI Agents can supercharge your company’s profits and productivity at TMC’s AI Agent Event in Sept 29-30, 2025 in DC.
Rich Tehrani serves as CEO of TMC and chairman of ITEXPO #TECHSUPERSHOW Feb 10-12, 2026 and is CEO of RT Advisors and is a Registered Representative (investment banker) with and offering securities through Four Points Capital Partners LLC (Four Points) (Member FINRA/SIPC). He handles capital/debt raises as well as M&A. RT Advisors is not owned by Four Points.
The above is not an endorsement or recommendation to buy/sell any security or sector mentioned. No companies mentioned above are current or past clients of RT Advisors.
The views and opinions expressed above are those of the participants. While believed to be reliable, the information has not been independently verified for accuracy. Any broad, general statements made herein are provided for context only and should not be construed as exhaustive or universally applicable.
Portions of this article may have been developed with the assistance of artificial intelligence, which may have contributed to ideation, content generation, factual review, or editing.
