Meta might be about to drop a cool $10 billion on Scale AI — and if it happens, this could be one of the biggest AI power plays we’ve seen yet. For context: Scale AI isn’t just any startup. It’s the behind-the-scenes MVP helping giants like OpenAI and Microsoft train their models by handling the not-so-glamorous but crucial task of data labeling.
And now? Meta wants in — big time.
While Meta has long championed its own open-source AI models (think Llama 3) and built much of its infrastructure in-house, this rumored investment signals a significant shift in its approach. Mark Zuckerberg has already teased plans to spend $65 billion on AI in 2025, and Scale might just be where a significant portion of that budget is allocated. Why? Because if you want smarter AI, you need great training data, and Scale is the go-to name for organizing the digital chaos.
Scale AI isn’t exactly a stranger to Meta, either. The two already teamed up on Defense Llama, a security-focused version of Llama 3 that’s helping military analysts and planners process high-stakes info. So, this is more than a strategic partnership — it’s a full-on alliance in the making.
Founded by Alexandr Wang at just 19, Scale AI has been on an absolute rocket ride. It raked in $870 million in 2024 and expects to more than double that in 2025. Despite some past controversy around labor practices (which, by the way, the company cleared without penalties), investors are still lining up — and Meta seems ready to make its biggest bet yet.
If this deal closes, we’re looking at a huge AI power shift. Microsoft has OpenAI. Amazon is backing Anthropic. And now Meta might be securing its data-labeling powerhouse in Scale AI. This could supercharge Meta’s race to dominate the AI landscape — and in a world where compute and clean data are the new oil, this move could give them one serious edge.
