What's After Meta's 49% Ownership in ScaleAI, 377% RAG Growth: The Next 12–18 Months in AI Data Services

After Meta’s 49% stake in Scale AI, investors are increasingly asking what other adjacent AI data service sub-segments could drive growth or margin expansion. Meta stake has sharpened focus not only on neutrality and concentration risk but also on the pathways for Scale AI and other companies to extend beyond its current core and build a broader moat in the evolving data-for-AI ecosystem.

The last 12–18 months have delivered some signals. The open-web free-for-all is over. As we noted in our previous article, OpenAI’s landmark deals with News Corp are now a strategic moat. Budgets are moving from ad‑hoc red teaming to programmatic evaluation‑as‑a‑service with regulator‑ready audit trails. In a recent development, Anthropic and the U.S. government's nuclear experts have developed a new tool that can spot the difference between a scientist asking Claude about nuclear reactors and a spy probing it for secrets about weapons development.

OpenAI and the UK government have signed strategic partnerships.

The rise of agentic AI—autonomous, tool-using systems like Microsoft Copilot agents and OpenAI Realtime agents—is changing what counts as “gold” in this industry: traceable interaction data and multi-turn evaluations.

We’ll also take a closer look at near-term plays from Scale AI and Innodata.

With all of these developments shaping the market, today’s article addresses the questions we are hearing most frequently from clients and investors:

1. Who’s really buying AI data today?
   Review 2024 State of AI Reports findings and understand market dominance for data

2. Which services matter most?
   RAG surge data & within the AI data services stack, what sub-segments are proving most interesting to clients and investors?

3. Who’s raising capital?
   Several AI data companies have raised significant rounds recently. Which ones stand out, and what’s the equity story behind them?

4. What’s the short and long-term trajectory?
   What does “forward-deployed capability” actually look like in practice?

   What long-term shifts should we expect beyond forward-deploy?

5. What is the growth outlook for the mid-market segment spending $50K–$500K annually?

   Review DataBricks AI Report and understand Scale AI’s opportunities

6. What are platform-control, on-premise trends?

   1. Are enterprises pushing to use their own annotation platforms, or are they adopting proprietary data platform?

   2. Are clients asking for on-premise annotation teams, embedded experts, or dedicated secure centers instead of purely cloud-based delivery?

Please note: The insights presented in this article are derived from confidential consultations our team has conducted with clients across private equity, hedge funds, startups, and investment banks, facilitated through specialized expert networks. Due to our agreements with these networks, we cannot reveal specific names from these discussions. Therefore, we offer a summarized version of these insights, ensuring valuable content while upholding our confidentiality commitments.