The AI Race Is Moving Into the Data Moat

The latest contest in artificial intelligence is not taking place only in model rankings or chip orders. It is moving into the organisations that possess the richest records of real-world decisions.

CrowdStrike’s launch of a Cyber Superintelligence Lab is a useful marker. The cybersecurity company is bringing AI researchers, threat hunters, offensive operators and incident responders into one research operation. Its argument is commercial as much as technical: a model trained against verified attacks may have a clearer route to value than a general system searching for another benchmark win.

“Security is how AI scales,” said George Kurtz, CEO and founder of CrowdStrike. “The Cyber Superintelligence Lab concentrates the PhDs, AI researchers, and the threat hunters who stop real attacks every day on the Falcon platform.”

CrowdStrike says its platform processes trillions of security events each day and carries fifteen years of threat intelligence and incident-response experience. That data is not simply large. It is labelled, investigated and tied to outcomes. In a market where customers are becoming more careful about AI bills, the distinction matters. Enterprises will pay for systems that reduce loss, speed decisions or prevent a breach. They are less likely to keep paying for capability that remains difficult to measure.

The infrastructure economy is still expanding at remarkable speed. NVIDIA reported quarterly revenue of $96.2 billion in late August, with data-centre revenue reaching $89 billion, up 117 per cent from a year earlier. The figures show where the present profits sit: with the supplier of the scarce compute that every major lab, cloud provider and specialist developer still needs.

Yet scarcity creates an opening for a different kind of competitor. Anthropic, OpenAI, Google, Meta and xAI are competing to make frontier models cheaper and more useful. Specialist companies are asking whether proprietary data, trusted workflows and a direct line to business outcomes can provide a stronger commercial position than scale alone.

That does not make the general-purpose labs less important. It changes the map around them. The AI race is becoming a contest between model makers, infrastructure owners and data-rich operators. The winners may be the companies that connect all three without asking customers to carry the entire cost.

Why Does It Matter?

For investors, the question is shifting from who has the most capable model to who owns the evidence that a model works. For executives, the practical test is harder: can AI improve resilience, revenue or judgement in a way that can be audited? The next layer of market power may belong to firms that can answer yes, with proof.

Get the stories that matter to you.
Subscribe to Cyber News Centre and update your preferences to follow our Daily 4min Cyber Update, Innovative AI Startups, The AI Diplomat series, or the main Cyber News Centre newsletter — featuring in-depth analysis on major cyber incidents, tech breakthroughs, global policy, and AI developments.