Answering Governments’ Toughest Intelligence Questions: Key Takeaways from the NatSec Tech Podcast with our CEO Jaap van Etten

Open-source intelligence (OSINT) plays an increasingly critical role in shaping export control decisions, investment screening, and broader national security policy. But as its use expands, so do the questions around its credibility, application, and coordination. These are questions we often hear from policymakers and analysts alike: 

• How can we be sure the data is accurate? 

• Is this work not already being done by intelligence agencies? 

• And if all the information is public, what is actually new? 

These same concerns were raised during a recent episode of the NatSec Tech podcast, where our CEO, Jaap van Etten, joined host Jeanne Meserve for a live conversation at the SCSP AI+ Expo in Washington, D.C.

While the interview also covers the story behind Datenna’s founding, something many of our readers may already be familiar with, we have chosen here to focus on a different set of takeaways: the questions governments are still wrestling with, and the answers that emerged from the conversation. 

OSINT Is Not Just Public, It Is Buried 

A common misconception about OSINT is that, because the information is technically public, it must be easy to find. In reality, much of it is hidden deep within regulatory filings, government databases, procurement platforms, academic pages, or outdated local websites. 

Jaap van Etten described how we solve this challenge by building what he called “10,000 mini apps”, highly specialized crawlers that work in parallel, scanning the web across China using VPNs and proxies to extract and reconstruct fragmented data. 

“The real work is not scraping a homepage. It is stitching together thousands of obscure data points, lab affiliations, fund disbursements, corporate links, into something coherent and explainable.” This work goes beyond surface-level search and requires ongoing maintenance, especially in a landscape where China is increasingly restricting access to sensitive information. 

Trust Comes From Traceability and Deep Expertise 

If access is one hurdle, trust is another. How can governments rely on OSINT if they cannot verify the source or context? Our approach is to treat every data point as a verifiable object. Each fragment is timestamped, linked to its original source, tagged with a trust score, and connected to the algorithm used to retrieve it. 

But technical traceability alone isn’t enough. That is why we employ a dedicated team of more than 25 China experts. These are analysts with deep knowledge of the Chinese science and technology landscape, the language, and the policymaking context. They play a key role in evaluating the reliability of sources and interpreting what’s behind the data.

“We tell our users: do not take our word for it. Use the source. Decide for yourself how much weight it carries.”

This combination of transparent data and contextual analysis ensures that governments can use open-source material with the same level of scrutiny they apply to classified intelligence. 

Why Are Governments Not Doing This Themselves? 

The host of the podcast asked a question we often hear: Why are intelligence services not already doing this? The answer is not a matter of capability but of focus. Government agencies are increasingly investing in OSINT, but many still face structural silos, limited cross-border visibility, and bureaucratic constraints that make long-term iteration difficult.

We have spent over a decade doing one thing: building a single knowledge graph focused on China’s science, tech, and military ecosystem. This depth, and the ability to work across allied governments, allows for faster iteration and a more complete picture.

“It is not that we were smarter,” Jaap van Etten said. “It’s that we started earlier and stayed focused.” 

OSINT Is Not Immune to Deception  

Of course, one strategic question often follows: Could the Chinese not just mislead you? 

Jaap van Etten’s answer was simple: “That is always possible, and we assume that’s the case.”

Just like traditional intelligence can be compromised, OSINT must be treated with caution. That is why redundancy, cross-referencing, and source-level transparency are built into our system. Not just to present information, but to allow analysts to assess its reliability.

“What we do with OSINT is what governments have always done with spies: validate, compare, and judge whether to act.”

Meanwhile, China is not just aware of OSINT, they are investing in it. Jaap van Etten estimated that “50 to 80 companies in China” are performing OSINT-like work, including some that track troop movements via social media and sell that data to the PLA. In parallel, the Chinese government is now restricting online discussions of military activity, fully aware of the intelligence value such data provides.

This arms race around open data is no longer theoretical. It is already happening. And it is one of the clearest signals that OSINT is not just useful, it is strategically consequential. 

How OSINT Is Shaping, or Missing, Policy 

Our data is already being used by governments for export control and investment screening. But Jaap van Etten emphasized that many systems are still overly focused on ownership, missing subtler, but more critical, links. For example: A company may be privately owned, but receive funding from a military-civil fusion initiative. Or be listed as a supplier to the PLA through a subsidiary or partner network.

“These kinds of links do not show up in a shareholder register, but they are precisely the ones that matter most.” To respond effectively, Jaap van Etten called for greater harmonization: of export controls, screening thresholds, and especially OSINT methodology. 

The Case for Harmonization: Building a Shared OSINT Backbone 

Our vision is not just better data, it is shared infrastructure. Too often, countries build their own OSINT systems in isolation. But if each government uses different standards and frameworks, their insights cannot be combined or compared. Jaap van Etten described a more collaborative future: 

• Data structured like “Lego blocks,” easily reused across agencies. 

• Built-in data lineage to avoid duplication or blind spots. 

• Shared access to verified fragments, even if analysis remains sovereign. 

“We need interoperability, not uniformity. That is how you scale trusted intelligence across borders.” 

The Path Forward 

The insights shared by Jaap van Etten highlight a critical truth: the future of national security hinges on how effectively allied governments leverage open-source intelligence. With Datenna, we stand at the forefront of this evolution, transforming fragmented public data into actionable intelligence and advocating for a more collaborative, proactive approach to protecting critical technologies. By providing unparalleled traceability, deep expertise, and a commitment to shared understanding, we empower nations to move beyond reactive measures and build a collective defense that truly matches the scale and ambition of today’s geopolitical landscape. 

Note: This article doesn’t recap the full podcast, which also includes Jaap van Etten’s personal story and broader reflections on tech competition and future collaboration. We’ve chosen to focus on the strategic questions governments are asking today. For the full conversation, listen to the NatSec Tech podcast.