SOC | About | News

When Critical Systems Fail: A New Test of National Resilience

Written by SOC News | Nov 20, 2026, 9:54:01 PM

By Shawn Roslin, Vice President of National Intelligence Division and International Programs

This past week’s disruptions highlight why infrastructure security must include continuity, redundancy and operational readiness.

Modern conflict is evolving faster than our preparations. Today’s emerging battlefields are digital, silent and without borders, where cyber warfare can be waged with a single keystroke from virtually anywhere. We often say that the next battle may not begin with explosions, but with a silent strike: a data breach, a blackout, or a system compromise. It could take days or weeks to understand the source of the attack. If we continue to train solely for yesterday’s wars, we risk being blindsided by tomorrow’s threats.

Global leaders in AI and defense understand the danger of uninvited access and surveillance by adversaries, knowing that some of the most potent weapons now travel through fiber-optic networks and cloud systems as advanced cyber tools engineered to disrupt and destabilize critical infrastructure.

Yet critical systems do not have to be deliberately attacked to create national consequences. Technical failures, maintenance errors, compromised identities, and interruptions to trusted services can produce many of the same operational effects

Recent disruptions are providing us with an alarming wake up call to illustrate how quickly system instability can affect essential operations:

  • FAA: A nearly five-hour nationwide ground stop affected 17 of 22 Air Route Traffic Control Centers and contributed to approximately 12,462 canceled flights. Travelers were stranded nationwide, flight schedules were disrupted and airlines faced days of crew, aircraft and passenger recovery challenges. Read the FAA’s incident summary.

  • United Airlines: A separate technology outage halted departures across the airline’s network for approximately 90 minutes, preventing aircraft from leaving until required flight information became available. Read the Associated Press report.

  • NSA: Scheduled maintenance degraded undisclosed internal systems supporting national-security operations. The agency did not disclose the duration, systems involved or number of users affected. Read the NSA’s statement on the network issues.

While public reporting has confirmed these incidents are not connected or attributed to cyberattacks, there is a valuable lesson and demonstration of what is possible. Whether the cause is malicious activity, aging infrastructure, a failed configuration or routine maintenance, the operational consequence can be similar: when a mission critical system or infrastructure people rely on suddenly becomes unreliable.

That is the larger security challenge. Critical infrastructure does not have to be hacked to create a national disruption.

 Disruption & Destabilization of Critical Systems  

The next crisis may begin with a failed router, an overloaded network, an unavailable cloud service, a compromised credential, or an upgrade that produces an unexpected result. It may also begin through access that appears legitimate: an authenticated session, vendor connection, service account or automated identity that has been compromised without triggering traditional perimeter defenses.

For the public, the distinction between an attack and a technical failure may matter less than the outcome: the disruption and destabilization of systems and/or critical infrastructure.

The systems supporting transportation, energy, communications, public safety, defense, and emergency response are no longer isolated. They are connected through data centers, cloud platforms, fiber networks, industrial control systems, software interfaces and third-party providers.

That interconnectedness creates enormous capability and a larger attack surface.

Industry forecasts indicate that this attack surface increasingly includes trusted identities and third-party connections. In its 2026 Forecast Report, SpyCloud found that 48% of surveyed organizations experienced account takeovers despite multifactor authentication, while 68% were affected by vendor or software-as-a-service breaches. The findings reinforce broader industry guidance: organizations must account for the employees, devices, vendors and automated identities operating through trusted connections, not only the systems protected by the traditional network perimeter.

Cybersecurity and physical security can no longer be treated as separate disciplines. A compromised network can affect physical operations. A damaged fiber route can interrupt digital services. An exposed data center can become a national-security concern. A disruption at one organization can create cascading effects for partners, customers, communities and the public.

As AI becomes embedded in all operations, the risks grow. AI models and training data can be stolen. Automated systems can amplify inaccurate information or accelerate a failure. A disruption in the data, communications, power or computing environment supporting an AI-enabled mission can quickly become an operational problem.

SpyCloud forecasts that stolen sessions, tokens and vendor connections will continue to cause much of the near-term damage, even as fully autonomous attacks receive more attention. Its report also points to AI-assisted social engineering and attacks targeting the credentials used by AI agents. The practical guidance is to incorporate AI into identity security, vendor oversight, and continuity planning rather than treat it as a distant or separate risk.

Resilience Requires Capacity & Redundancy

The challenge is unfolding as the federal capacity responsible for reducing cyber risk is under pressure.

CISA and the FBI play complementary roles in the national cyber-defense ecosystem. CISA supports civilian cyber defense, critical-infrastructure coordination and information sharing. The FBI leads investigations involving criminal cyber activity and national security.

Public budget documents and congressional testimony indicate that federal infrastructure-security capacity is under pressure. The Department of Homeland Security’s fiscal year 2026 budget justification proposed a workforce-transition reduction involving 325 positions, 315 full-time equivalents and $64.9 million at CISA. Congressional testimony subsequently described the agency’s active workforce as having declined by approximately one-third, although the figures vary depending on whether they include authorized positions, vacancies, voluntary departures, transfers or active employees. Read CISA’s fiscal year 2026 congressional budget justification and the related congressional testimony.

The Government Accountability Office also reported that active CISA personnel dedicated to chemical-sector activities declined from 214 in fiscal year 2024 to 52 in fiscal year 2025. The report described fewer on-site facility assessments and reduced engagement between CISA and some private-sector stakeholders. These findings do not establish that staffing changes caused any recent disruption. They do, however, raise questions about the capacity available for infrastructure assessments, information sharing and incident coordination as the risk environment becomes more complex. Read the GAO report on CISA’s chemical-security responsibilities.

Congress is responding on several fronts. Following the November 13 outage, Sen. Eric Schmitt introduced the Modern Skies Act, a proposed $30 billion effort to accelerate replacement of aging air-traffic-control equipment. Senate Commerce Committee Democrats also requested technical briefings and clear, fact-based answers about the outage, including why existing redundancy did not restore service immediately. Congress has also continued cyber-threat information-sharing authorities and is considering proposals to use AI to identify and remediate vulnerabilities in critical infrastructure.

Resilience cannot begin after a breach. It must be designed into the mission from the start.

That means modernizing aging infrastructure, protecting operational technology, segmenting networks, securing facilities and data centers, maintaining redundant communications and power, and preserving the ability to operate when automated systems are unavailable. It also means testing contingency plans under realistic conditions and ensuring that public and private partners can share information quickly when something goes wrong.

Organizations must also understand which employees, contractors, vendors and automated identities have access to critical systems and how quickly that access can be revoked. Public communication should be included in continuity planning as well. An outdated status page or inconsistent message can compound a technical disruption by creating uncertainty about which systems remain operational and what the public should do next.

Security is not simply the prevention of unauthorized access. It is the preservation of mission availability.

Recent events should remind us that resilience is measured in the moment systems are stressed, not when everything is functioning normally. The organizations best prepared for the next disruption will be those that understand the full operating environment: the facility, the network, the data, the people, the supply chain and the mission they are built to support.

For Day & Zimmermann Government Services, that integrated view is central to the work of SOC and Mason & Hanger. Secure facilities, reliable infrastructure, physical protection, cyber defense and operational continuity must work together. In an increasingly connected environment, protecting one system means understanding the systems around it and preparing for what happens when one of them fails.

The systems we depend on will be tested. Our responsibility is to ensure that the mission continues when they are.