Key Takeaways

  • Sovereign AI models are less about nationalistic tech stacks and more about nurturing local talent and research to create domestic demand for compute, especially in regions like Europe.
  • Building essential AI infrastructure, like data centers, means facing public resentment and regulatory delays, which Nebius counters by maintaining "oversubscribed" project capacity.
  • Community engagement for data center construction is no longer optional PR; it’s a "duty to educate" local authorities and citizens on project value and impact.
  • The growth of open-source AI models, counterintuitively, fuels an increasing appetite for specialized compute, making the data center build-out even more urgent and contentious.

The Unseen Fight for AI Autonomy

The idea of "sovereign AI" often conjures images of nationalistic tech stacks. But for Roman Chernin, co-founder of Nebius, the true battle for AI autonomy plays out far from the code. It starts with demand. Harry Stebbings, host of 20VC, pressed Chernin on the geopolitical lines shaping AI: "How important do you think it is that nations have their own sovereign models? It looks like the world is divided or we may not like it." Chernin’s answer cut through the rhetoric.

For him, the focus should not just be on the models themselves, but on nurturing the very ecosystem that uses them. “What we need to care about here is to have more great companies like lovables, black forest labs, I don't know, mist drives of the world and we have enough people that invest in research, have enough people that invest in the products.” He argues that regions like Europe need to develop “good enough foundational models” locally, not just to avoid reliance on foreign tech, but to spark a local hunger for compute. This demand, generated by local builders and researchers, becomes the engine for infrastructure development. Without an active, hungry domestic market, the "sovereign" part of sovereign AI is just a hollow shell. This is a critical insight: the path to national AI strength isn't through government mandates, but through fostering a vibrant, hungry domestic market.

Data Centers Are Built on Diplomacy

Once that demand exists, the next fight begins: physically building the compute. Data centers, the silent giants of AI infrastructure, face intense public resentment. They consume vast amounts of land, power, and water, often drawing the ire of local communities and environmental groups. For Nebius, a company deeply involved in full-stack AI infrastructure, navigating these challenges isn’t a side quest; it’s core business. Chernin laid out their pragmatic approach to this battlefield. Delays are inevitable, so Nebius plans for them.

“We need to make sure that we are like overs subscribed if you want and if one data center will be delayed we will still deliver enough capacity to our customers.” This requires smart project management. It also recognizes the political realities of infrastructure development. Chernin sees community engagement as a non-negotiable duty. “Communities and the local authorities require the companies like us to work closely with them and explain and show what what what we do and work with them on their concerns and like address them. This is the reality.” For Chernin, founders cannot hide behind technical jargon. They must proactively educate, listen, and collaborate, turning potential resistance into at least grudging acceptance, if not active support. This proactive diplomacy is as essential as the engineering itself.

What to Do With This

If you're building any large-scale infrastructure, or even just planning a move that might draw public attention, adopt Chernin's "oversubscribed" mentality. Immediately identify your project's most contentious points for public or regulatory pushback (energy, land use, traffic) and map out at least two backup scenarios for potential delays. Then, draft a simple, jargon-free message explaining your project's benefits to a local community group and plan your first outreach this week.