At a time when global security feels increasingly unpredictable, one message from NATO leadership is coming through clearly: the intelligence advantage won’t come from collecting more data. It will come from using it faster—and better.
Speaking at GEOINT 2026, Major General Paul Lynch laid out a candid assessment of where allied intelligence stands today—and where it needs to go next.
The conversation, moderated by USGIF CEO Ronda Schrenk, wasn’t theoretical. It was grounded in real-world lessons from Ukraine, evolving threats from Russia and China, and the growing role of commercial and AI-driven intelligence.
The takeaway? The future of allied intelligence isn’t about more sensors. It’s about fixing the pipeline from collection to decision.
A More Dangerous—and Faster—Security Environment
The backdrop to Lynch’s remarks is a rapidly shifting global landscape.
Russia’s continued war in Ukraine, combined with support from China and North Korea, has reshaped NATO’s strategic posture. Add in cyberattacks, disinformation campaigns, and economic coercion, and the alliance is facing what Lynch described as a “much harsher strategic environment.”
At the same time, NATO is scaling up:
- Defense spending across European allies and Canada jumped roughly 20% in 2025
- The alliance has committed to a 5% GDP benchmark by 2035
- Force readiness, forward presence, and operational planning are all increasing
But Lynch made one thing clear: more spending alone doesn’t equal better security.
The Ukraine Lesson: Speed Beats Perfection
If there’s one place where modern intelligence has been stress-tested, it’s Ukraine. And the lesson isn’t subtle.
According to Lynch, the most important shift has been this:
- Speed matters more than perfection
- Fusion delivers advantage
- Sharing earlier is better than waiting for certainty
In practical terms, that means intelligence must move in hours—not days. When it works, it’s powerful. Lynch described allied intelligence support to Ukraine as some of the most sophisticated he’s seen in his career. But when it fails, the problem usually isn’t collection. It’s integration.
The Real Problem: The Pipeline Is Broken
For an industry built on satellites, sensors, and data collection, Lynch shared that intelligence failures are “almost never a collection failure. It’s an integration problem.”
The issue isn’t a lack of data. It’s everything that happens after:
- Processing and exploitation workflows
- Classification and releasability rules
- Data-sharing frameworks across 32 allies
- Getting intelligence into a usable format for decision-makers
This is what he called the “pipeline from collection to decision.” And right now, that pipeline isn’t built for the speed of modern conflict.
The Commercial GEOINT Explosion
Another major shift: commercial space and GEOINT providers are no longer supplemental—they’re central.
Compared to early operations in Iraq and Afghanistan, today’s environment looks completely different:
- Higher revisit rates
- Better resolution
- Faster analytics
- More accessible data across allies
This is leveling the playing field. Smaller NATO nations can now contribute meaningfully to intelligence efforts without owning national satellite systems. NATO has even released its first commercial space strategy, recognizing this reality. But again, the challenge isn’t capability. It’s integration. Most commercial data still enters NATO systems through workarounds—not designed pathways.
AI Is Changing Everything—And Creating New Risks
Artificial intelligence is accelerating the problem. AI can dramatically reduce the time between collection and actionable intelligence. It enables automated imagery analysis, faster multi-source fusion, and scalable pattern detection. But it also introduces a new layer of complexity—especially in a 32-nation alliance.
Lynch raised a very real scenario: Two different AI models, trained on different national datasets, produce conflicting intelligence reports. Now what? Which model is trusted? What data trained it? What confidence level is acceptable?
This isn’t a future problem. It’s already happening. And it highlights a critical gap: AI governance.
NATO’s Biggest Challenge Isn’t Technology—It’s Governance
If there’s a theme running through Lynch’s remarks, it’s this: The biggest constraint isn’t technical. It’s governance.
That includes:
- Who can share what
- With whom
- Under what authority
- And at what speed
Across 32 nations, each with its own classification systems, legal frameworks, and policies, this becomes incredibly complex.
But it’s also where the biggest gains can be made.
Lynch suggested that redesigning these frameworks could be one of the highest-return investments NATO makes in the next five years.
What NATO Needs from the GEOINT Community
Lynch closed with a direct call to action—specifically for the GEOINT community. He outlined three priorities.
1. Sharing Must Become the Default
Current systems treat sharing as an exception. That has to flip.
Modern operations require intelligence to move quickly across allies without manual escalation.
2. Data Standards Must Catch Up
Interoperability isn’t optional anymore. There needs to be common metadata, standardized AI model documentation, and shared data formats. Without this, every integration becomes a custom solution—and a delay.
3. Relationships Matter More Than Policy
In some of the earliest days of the Ukraine conflict, intelligence sharing moved faster than policy allowed. Why? Because people trusted each other.
Years of exercises, exchanges, and collaboration created the relationships that allowed decisions to happen quickly—even before formal frameworks caught up. Lunch shared, “No policy document produces that. Relationships do.”
