WarBrief Live | October 4, 2026 | International Security
Modern conflicts are fought with missiles and drones, but they are mapped with data — and some of the most revealing data is hiding in plain sight. Every day, internet-wide scanners such as Censys systematically probe the public internet and publish what they find: which industrial controllers are reachable, which building systems answer to strangers, which certificates tie infrastructure together. This freely available scan data has quietly become a pillar of Censys threat intelligence work: researchers, journalists, and security teams use it to illuminate the exposed infrastructure that underpins economies, militaries, and critical services — the same infrastructure adversaries study when planning cyber operations.
Key Takeaways
- Internet-wide scan data from platforms like Censys documents exposed industrial and building-control systems at global scale — Censys researchers have identified 1,500 exposed control systems across 18 automation protocols.
- A 2026 exposure analysis found more than 6,300 internet-reachable industrial control devices near 1,063 U.S. data centers, concentrated in building-automation protocols.
- This visibility is dual-use: defenders use it to find their own exposure, while adversaries can use the same public data for targeting — making exposure management a national-security issue.
- Scan data has limits: IP geolocation indicates proximity, not proof of placement, and responsible analysts corroborate before attributing infrastructure to any facility or actor.
What internet scanning reveals about the physical world
The public internet is not just websites. It is also the control layer of the physical world: programmable logic controllers running factories, building-automation systems managing data-center cooling, human-machine interfaces overseeing power distribution. Many of these systems speak industrial protocols — Modbus, BACnet, Siemens S7, DNP3 — that were designed decades ago without authentication, on the assumption they would never face the open internet.
That assumption has failed at scale. Censys’s own research into industrial control system exposure discovered 1,500 control systems exposed to the public internet using 18 automation protocols, alongside 1,700 publicly accessible HTTP devices associated with 26 operational-technology vendors — many of them web administration interfaces, a large share likely still using default credentials. Over 80 percent of the exposed interfaces managed building controls. Because these low-level protocols typically lack authentication, a reachable device is often a controllable device.
The picture keeps getting sharper. An internet-exposure analysis published in August 2026 identified more than 6,300 high-confidence industrial control and building-automation devices accessible near 1,063 U.S. data centers — the cooling, power-conditioning, and environmental systems that keep server halls alive. BACnet controllers accounted for 58 percent of the verified sample, with Niagara Framework instances adding another 23 percent. The researchers filtered nearly 74,000 raw records down to verified devices across 3,991 unique IP addresses, excluding honeypots and false positives.
Why conflict watchers care

Critical infrastructure has become a central theater of modern conflict — energy grids, logistics hubs, and communications networks feature in nearly every contemporary confrontation, from Ukraine to the Middle East. Cyber operations against such targets begin with reconnaissance, and internet scan data collapses the cost of that reconnaissance to zero. An adversary no longer needs to probe a target directly, generating suspicious traffic; the exposure map already exists, compiled by someone else’s scanners and queryable from anywhere.
This is the dual-use dilemma at the heart of Censys threat intelligence. The identical dataset serves two masters. Defensive teams and national computer-emergency response teams use it to discover their own country’s exposure and push owners to remediate — effectively a free, continuous national exposure audit. Threat actors use the same queries to build target lists. The data does not pick sides; the race is between those who remediate and those who exploit.
Intelligence analysts tracking conflicts use scan data differently again: as context. Pivoting from an IP address through certificates and network ownership — the methodology in our Censys Search investigations guide — can reveal infrastructure relationships that corroborate or challenge other reporting. When cross-checked against satellite imagery, shipping data, and on-the-ground reporting, exposure data becomes one more sensor in the intelligence picture. Compared side-by-side with Shodan’s device archive, it gains the redundancy that serious analysis requires.
The “Street View” principle — and its limits
Censys founder J. Alex Halderman has long argued the project merely organizes what is already public:
“It’s similar to Google Street View, where we’re gathering what’s already publicly visible and making it available in one place. To extend the analogy, we just take a picture from the sidewalk. We don’t peek in the door, we don’t jiggle the locks.”
The analogy is honest about method but incomplete about consequence. A photograph of a house does not usually include the state of its locks; a scan of an industrial controller effectively does. Publishing that a specific protocol with no authentication is reachable at a specific address is materially more useful to an attacker than a street photograph is to a burglar. This is why the research community pairs scanning with responsible disclosure norms: measuring exposure in aggregate, notifying affected owners where possible, and avoiding step-by-step targeting guides.
Analysts must also respect the data’s limits. The data-center study’s authors were explicit: IP geolocation provides metropolitan proximity, not proof that a given device sits inside a specific facility. Treating scan results as an attribution list — “this controller belongs to that power plant” — is how careful measurement curdles into misinformation. Corroboration is not optional; it is the difference between intelligence and speculation.
What states and defenders are doing about it

The policy response has three layers. First, exposure reduction: industrial protocols were never meant to face the internet, and every device pulled behind a VPN or access gateway is a device removed from the public index. National cybersecurity agencies now routinely cite scan-derived exposure figures when urging critical-infrastructure operators to segment operational technology from the public internet.
Second, continuous self-surveillance: mature security teams query scan platforms for their own address space the way they check credit reports — routinely, looking for surprises. If a scanner like the one behind the CensysInspect user agent can see it, so can everyone else.
Third, threat-informed defense: knowing which of your systems appear in public indexes tells you what an adversary’s cheapest reconnaissance already knows, letting defenders prioritize accordingly. Our response guide for scanner traffic in logs covers the operational side — what to do when the census-takers knock.
The strategic takeaway is uncomfortable but clarifying: in an era when the entire public internet is re-surveyed daily, security through obscurity is dead. Anything reachable is indexed; anything indexed is knowable. Nations and companies that internalize that — and design as if the adversary already has the map, because they do — will fare better than those still hoping nobody looks. For more on the tools behind this visibility, see WarBrief’s intelligence tools hub.
Frequently asked questions
How is Censys data used for threat intelligence?
Analysts query Censys’s internet-wide scan index to discover exposed infrastructure — reachable industrial controllers, misconfigured services, certificate relationships — and to track how exposure changes over time. Defenders use it to find their own vulnerable systems; researchers use it to measure national-level exposure; threat actors can abuse the same public data for targeting, which is why remediation speed matters.
What did Censys find about exposed industrial systems?
Censys research identified 1,500 control systems exposed to the public internet across 18 automation protocols, plus 1,700 publicly accessible HTTP devices from 26 operational-technology vendors. A separate 2026 analysis found over 6,300 internet-reachable industrial control devices near 1,063 U.S. data centers, mostly building-automation systems.
Can internet scanning data attribute infrastructure to a specific country or facility?
Only weakly on its own. IP geolocation indicates approximate regional proximity, not proof of physical placement or ownership. Responsible analysts treat scan data as a lead to be corroborated with other sources — never as standalone attribution.
Is it legal for Censys to scan the entire internet?
Censys scans only what systems publicly expose — it completes normal protocol handshakes and records the responses, a practice rooted in academic internet-measurement research. Networks can opt out of scanning. Laws vary by jurisdiction, but the measurement itself is widely practiced by researchers, companies, and governments.
How can organizations reduce their exposure in scan indexes?
Remove industrial and building-control systems from direct internet reach (VPNs, jump hosts, network segmentation), change default credentials, close unneeded ports, and continuously query scan platforms for your own address space to catch regressions.
Sources
- Cyber Press: Censys research on 1,500 exposed control systems and 18 automation protocols
- GBHackers: 6,330 internet-exposed ICS devices near U.S. data centers (August 2026)
- University of Michigan: Censys origins and the “Street View” measurement philosophy
- Wikipedia: Censys — scanning scale, ZMap lineage, and user base
- SANS Internet Storm Center: Censys scanning activity and methodology
Read next: Iranian Hackers Suspected in Widening US Water and Energy Grid Cyberattacks — how a suspected Iranian-affiliated cyber campaign moved from water utilities to energy grids and telecom networks across a dozen US states.
[…] scan internet addresses for controllers, dashboards and remote-access services — the same internet-scanning exposure that puts infrastructure behind conflict zones at risk; log in with a default or stolen password; or exploit an unpatched vulnerability. No advanced […]
[…] see whether awareness campaigns and advisories are actually working. This connects directly to our analysis of exposed infrastructure during conflict, and to the [[CLUSTER-9]] piece on military and government cybersecurity, where national-scale […]
[…] CensysInspect Explained: The Internet Scanner CensysInspect Explained: Internet Intelligence Guide, How Internet Scanning Exposes Infrastructure Behind Conflict, Censys Internet-Wide Scan Data for Security Research, Cyber Warfare: Russia-Ukraine Conflict […]
[…] a deeper grounding, read our guide to how internet scanning exposed infrastructure in conflict zones, our explainer on what Censys Inspect is, and the research-methods primer […]
[…] how internet scanning exposed infrastructure in the conflict for background on how scan data reshaped this kind of […]
[…] background reading on exposed infrastructure in conflict zones can be found in our earlier piece on internet scanning and exposed infrastructure in conflict, and the defensive research toolkit behind such analysis is described in Cyber Warfare Research: […]