Modern cybersecurity architecture begins with a curious assumption: trust is dangerous.
Security frameworks associated with John Kindervag operate from a simple premise—no user, device, or system should be trusted merely because it appears familiar. Every request must be verified and every interaction authenticated. Anyone who has spent time working in cybersecurity eventually notices how easily trust assumptions fail. Systems behave exactly as designed until the day someone discovers how those assumptions can be manipulated.
Yet this philosophy would likely have seemed strange to the engineers who built the early Internet. The architects of ARPANET worked within relatively small research communities where participants generally trusted one another1. Protocols were designed for cooperation rather than hostility.
Looking back, this reveals something about technology that engineers sometimes forget. Technical systems rarely exist apart from the social environments that produce them. The early Internet was built inside a culture of cooperation, and its architecture seems to reflect that culture. What later generations discovered is something societies have learned many times before: trust is not merely a virtue between individuals. It behaves more like infrastructure—and like any infrastructure, it can erode.
Trust as Social Infrastructure
Political scientist Francis Fukuyama explored a version of this idea while examining the role of trust in modern economies. In Trust: The Social Virtues and the Creation of Prosperity he argued that societies with high levels of interpersonal trust often build larger and more effective institutions2.
One begins to see why when looking at how organizations actually function. Complex institutions require cooperation among people who may never meet one another. They require individuals to rely on the competence and honesty of strangers. Without some degree of trust, every interaction would require exhaustive verification.
In such a world, institutions would slow to a crawl.
Fukuyama’s argument was directed primarily at economic life, but the same pattern appears in technological systems. Modern networks depend upon layered relationships of trust. When a browser connects to a website, it trusts a certificate authority to confirm the identity of the server. Software updates assume the integrity of developers and distribution channels. Even the routing of data across the Internet depends upon systems behaving as expected.
These relationships form a vast web of interdependence. Perhaps it is helpful to think of modern civilization as operating within something like an ecology of trust.
A Fragile Environment
Ecological systems are rarely stable in the way machines are stable. They consist of networks of interdependent relationships, and disturbances in one part of the system can ripple outward in unpredictable ways.
Digital systems seem to behave in much the same way.
A compromised software repository can distribute malicious code to thousands of organizations before anyone notices. A forged certificate can undermine encrypted communications across large portions of the Internet. A vulnerability in widely used software can spread quietly across entire industries.
The systems themselves are complex, but the fragility often lies elsewhere—in the relationships of trust embedded within them.
Some scholars have begun to describe modern conflict in these terms. As Thomas Rid observes in Active Measures, many modern influence campaigns aim less at destroying infrastructure than at undermining confidence in institutions3. The target is not always the system itself; often it is the trust that allows the system to function.
Once trust weakens, everything else becomes more difficult.
The Internet’s Innocent Beginnings
Students encountering early Internet protocols often react with disbelief. How could engineers design systems that allowed something as simple as email sender spoofing or unauthenticated routing updates?
The answer becomes clearer when one remembers the environment in which the network emerged.
The early Internet grew out of research communities connected through the Defense Advanced Research Projects Agency. Participants were universities and laboratories populated by scientists who often knew each other professionally. Within that environment, elaborate verification systems were unnecessary. Cooperation was assumed.
In retrospect, the architecture of the Internet seems to reflect the culture of the people who built it. Trust was embedded in the system—not because engineers were naïve, but because their world was smaller.
The Internet eventually grew beyond that world. When the environment changed, the trust assumptions embedded in the technology gradually became vulnerabilities.
When Trust Fails Quietly
Trust failures rarely announce themselves in dramatic fashion. More often they appear only in hindsight.
At this point a reader might reasonably ask what any of this has to do with everyday work inside technology companies. Discussions about trust can easily drift toward abstraction—institutions, societies, civilizations. Yet the fragility of trust is often easiest to see in ordinary professional environments, where systems depend quietly on assumptions about the people who operate them.
Around 2005, while working for Quantum3D performing verification and validation on simulation hardware and software platforms—and serving as security manager for the company’s cleared subsidiary, CG2 (now Quantum3D Government Systems)—I occasionally spoke with an engineer named Sheldon Meng. Our interactions were brief and infrequent; he worked in the company’s San Jose office while I was based in Huntsville.
One conversation I remember had nothing to do with engineering. We spoke about genealogy, a subject that has long fascinated me. Meng mentioned that his family claimed descent from a brother of Confucius. Whether the story was true I never learned, but the conversation stayed with me long after the incident—not because of what it revealed about him, but because of how easily trust grows in ordinary human interaction.
Some time later the company learned that proprietary source code had surfaced in a competing product being demonstrated overseas. The trail eventually led back to Meng. When he returned to the United States he was arrested, and investigators from the U.S. Department of Justice reportedly confirmed that the code on his laptop matched Quantum3D’s software4.
What struck me then—and still does now—was how ordinary the situation had appeared beforehand. Trust in organizations rarely fails through grand conspiracies. More often it fails quietly, through individuals who seem entirely unremarkable until the moment a trust assumption collapses.
Incidents like this suggest that trust is rarely a technical problem. It is usually a human one.
Ancient Authentication
Human societies have long wrestled with this problem. Long before networks and authentication protocols existed, communities needed ways to distinguish insiders from outsiders.
One of the earliest recorded examples appears in the Book of Judges. In the aftermath of a battle between the Gileadites and the Ephraimites, those attempting to cross the Jordan River were asked to pronounce a particular word: “shibboleth.” Those who could not pronounce it correctly were identified as outsiders5. The story reads almost like a primitive authentication protocol.
Communities have always required mechanisms for verifying identity and membership. Passwords, uniforms, dialects, and cultural knowledge have served as signals of belonging. Modern authentication systems—passwords, digital certificates, and multi-factor authentication—perform essentially the same function.
The technologies are new. The underlying problem seems remarkably familiar. The technologies change, but the human problem rarely does.
Engineering Distrust
As the Internet matured, engineers gradually recognized that implicit trust assumptions could no longer be sustained. Networks had become too large, too complex, and too exposed to hostile actors.
This recognition gave rise to new approaches to security architecture.
The concept of Zero Trust, articulated by John Kindervag, reflects this shift in thinking. Traditional network security often assumed that systems inside a trusted perimeter could operate with relative freedom. Once inside the network, users were generally trusted. Zero Trust abandons that assumption. Every interaction must be authenticated, every request evaluated, and every device verified.
From one perspective, this looks like an engineering response to a changing environment. Verification becomes a substitute for assumption.
The Fragile Ecology
After enough years in cybersecurity, one begins to notice that many security problems are not really technical failures. They are trust failures that technology merely exposes.
Engineers designing Zero Trust networks are confronting a problem older than the Internet itself: how to maintain cooperation among strangers in environments where deception is possible. Ancient communities devised passwords like the “shibboleth.” Political revolutions questioned the legitimacy of inherited authority. Modern digital infrastructure encodes similar concerns into authentication protocols and verification systems.
The tools have changed, but the underlying challenge remains surprisingly consistent.
Modern civilization appears to run on fragile networks of trust—between institutions, experts, and the systems that mediate everyday life. Cybersecurity professionals therefore find themselves engaged in work that extends beyond defending networks. In a quiet way, they help maintain the conditions that allow complex societies to function.
Perhaps that is why modern security architecture begins with such an unsettling premise. Trust can be dangerous. Yet without it, civilization itself would become impossible—and every system we build would eventually collapse under the weight of verification alone.
References
- Defense Advanced Research Projects Agency. (n.d.). ARPANET and the origins of the Internet. https://www.darpa.mil ↩︎
- Fukuyama, F. (1995). Trust: The social virtues and the creation of prosperity. Free Press. ↩︎
- Rid, T. (2020). Active measures: The secret history of disinformation and political warfare. Farrar, Straus and Giroux. ↩︎
- U.S. Department of Justice. (2008, June 5). Former engineer charged with theft of trade secrets from Silicon Valley company. https://www.justice.gov/archive/opa/pr/2008/June/08-nsd-545.html ↩︎
- The Holy Bible, Judges 12:5–6 (English Standard Version). https://www.esv.org/Judges+12/ ↩︎
