Cyber Experts Mostly Blame Interconnected Systems and AI ‘Patching’ For Outages, But Warn Cyber Attacks Remain a Threat
Cybersecurity experts say recent U.S. outages point to underlying systemic vulnerabilities that expose the dangers of greater interconnectedness as well as the risks posed by AI and its ability to identify flaws at speed. The experts also warn that cyber attacks remain a danger.

In the wake of a series of major technology outages which have affected major federal agencies and private companies, speculation has grown about the origin of these failures, which have caused major disruption to U.S. aviation, banking, national security communications and essential digital services.
National security officials say that, while Iran and other foreign entities have used cyber attacks to target key infrastructure in the past, they are not behind these tech outages. Department of Homeland Security Secretary Markwayne Mullin told Fox News that the cases are "individually unrelated" and claimed that "one guy that is very smart with tech" had told him that SI (super intelligence), the name the Trump administration uses to refer to artificial intelligence, was behind the string of stoppages.
But what do the experts say?
Nigel Gibbons is a director and senior advisor at UK-based global cybersecurity and cyber resilience company NCC Group. Gibbons spoke to International Business Times about his diagnosis of the recent tech problems that Washington has been experiencing.
Gibbons suggested that, while there may no be explicit relationship between the individual cases, a heavily modernized and interdependent patchwork of systems can have a domino effect, meaning that one tech issue can quickly become several.
"Systems and services that appear entirely independent may actually rely on the same underlying technology providers, infrastructure and suppliers. This interconnectedness is one of the defining systemic vulnerabilities of modern digital infrastructure ... a problem affecting one component can quickly disrupt other systems that depend on it," Gibbons told IBT.
"While recent outages should not necessarily be viewed as evidence of an immediate systemic crisis, they reinforce the importance of understanding how disruption in one part of the digital ecosystem can affect services across multiple industries," he added.
Although Gibbons clarified that "not every major outage is caused by a cyber attack," his colleague Matt Hull, who works as vice president of Cyber Intelligence and Response at NCC Group, warned that the downstream effects of these tech outages create openings for malicious cyber actors.
"Cyber criminals move quickly to exploit the uncertainty, disruption and frustration that events like this create," Hull said to IBT.
Hull's analysis may prove particularly relevant in relation to the banking service interruptions. On Monday, JPMorgan Chase experienced a software failure which rendered thousands of customers unable to access banking services at a variety of institutions.
Three days later, JPMorgan confirmed that it had been the victim of a cyber attack which may have enabled hackers to access customers' names and email addresses. The bank, however, has insisted that the hack was unrelated to Monday's incident, but the incidents nevertheless demonstrate how technical disruptions and cyber threats can occur nearly simultaneously.
Robbie Mueller, a senior solutions engineer at California-based cybersecurity ArmorCode, also stressed that the recent problems can be explained without pointing the finger at nefarious actors.
"Nearly all of these [recent incidents] point to a change the organization made itself, or to load hitting something old. None of them requires an attacker."
This seems to have been the case with regards to the incident which affected U.S. flights. Transportation Secretary Duffy attributed the Federal Aviation Administration's (FAA) decision to halt all departing flights for four hours on November 13 to a telecommunications issue that came about after a government contractor installed a new router as part of a wider "analog to digital" upgrade of the system.
What about AI?
There have been multiple recorded instances of AI agents belonging to major companies such as OpenAI and Anthropic going "rogue," circumventing safety guard rails and subsequently causing data and security breaches while circumventing. This, however, does not appear to have caused recent issues.
Instead, AI's role in the crises may stem from its advanced ability to identify vulnerabilities and potential flaws in existing technology at breakneck speed which, in the words of Democratic Senator Mark Warner, creates "enormous pressure on companies and government agencies to patch those vulnerabilities immediately, often all at once, putting tremendous strain on systems that weren't built to handle that pace of change."
Ramy Rahman, a senior principal solutions engineer at ArmorCode, agrees with Warner, identifying AI, and its "speed," as the main culprit.
"There is definitely a large amount of fragile legacy infrastructure that is probably going to fall over if they get patched with the speed of AI ... disruptions are happening because of AI. That's about as simple as it gets," Rahman said to IBT in response to a question about the existence of common technical patterns or vulnerabilities that characterize all the events.
Daniel Miessler, a San Francisco-based AI and cybersecurity researcher, agrees.
"I think it [AI] makes them [the disruptions] much more likely, even in cases where no AI agent touched the system that went down," Miessler told IBT.
"AI is changing the whole environment around these systems: more attackers, more vulnerabilities being found, more scanning, and more pressure to patch fast. So when a company says no AI agent was involved in its outage, that can be accurate and still leave AI as a big part of why these are happening," the researcher continued.
This notwithstanding, Miessler also pointed out that, while it may be true that neither rogue AI models nor state actors are behind the series of shutdowns and interruptions, the risk of AI-enabled cyber attacks is very real.
"AI models and openly available offensive tools are making attackers a lot more dangerous ... Iranian groups have been going after US critical infrastructure for years, including water utility control systems and brute-force campaigns against healthcare, government, and energy networks, and they're getting the same AI uplift as everyone else," he warned.
© Copyright IBTimes 2026. All rights reserved.

