Field report S-43 / Q1 · sheet /operations
Operations & RestructuringFiled from the territory
One Router, 18,000 Cancelled Flights: America's Aging Tech Fails
A week of failures at Chase, Microsoft and the FAA traced back to aging infrastructure under upgrade strain — not cyberattacks. One router grounded 18,000 flights.
- Grid
- S-43 / Q1
- Surveyed
- November 18, 2026
- Reading
- 3 min read
- Correspondent
- Sophie Lindqvist

Field notes
- A single router installed at an FAA facility in Arkansas caused a ground stop that cancelled more than 18,000 flights over three days.
- Third-party tech providers caused roughly two-thirds of publicly reported IT outages over the past nine years, per Uptime Institute.
- Microsoft attributed its outage to a fault during a network capacity upgrade.
- Homeland Security Secretary Markwayne Mullin said the disruptions were not tied to cyberattacks.
A single router installed at a remote FAA facility in Arkansas triggered a nationwide ground stop that cancelled more than 18,000 flights over three days last week. That figure, reported by Reuters, is the starkest measure of a five-day stretch in which Americans struggled to send emails, board planes and reach their bank accounts across failures at JPMorgan Chase, Microsoft, the FAA and other major institutions.
The emerging consensus among officials and industry analysts is uncomfortable but straightforward: this was not a coordinated cyberattack. It was the sound of corporate America's technology debt (the accumulated cost of maintaining outdated systems instead of replacing them) coming due all at once.
What caused the outages?
Marva Bailer, CEO of tech advisory firm Qualaix, told The New York Post that outages of this kind often strike while systems are undergoing major upgrades. Bolting new technology onto aging infrastructure leaves the combined system vulnerable, she said.
Microsoft has already confirmed that pattern in its own case. The company attributed its outage to a fault that arose during a network capacity upgrade, according to the Financial Times. It remains unclear whether the incidents were directly AI-related — a caveat worth holding onto — but many organizations are presently reworking their networks to accommodate AI tools, which adds upgrade pressure precisely when systems are least stable.
U.S. Homeland Security Secretary Markwayne Mullin offered the same diagnosis on CNBC. "[They're] pushing new technology on top of really old computer systems, and it just seems to overwhelm the system," he said, in comments reported by Reuters. When a cabinet-level official and a startup advisor converge on the same explanation, skepticism about more dramatic theories — state-sponsored hacking, AI-driven assault — loses ground.
Why now, of all times?
Timing matters. The failures clustered in mid-November, when technicians rush to prepare systems for the holiday shopping and travel season.
"It's holiday time, it's travel time, it's shopping," Bailer told The New York Post. "So, you have all those higher spikes in volume and more pressure to [update your systems]."
There is a second, quieter dynamic at work. As The Post reported, AI tools now let companies detect software vulnerabilities far faster than before. That is theoretically good news — and practically a bottleneck, because IT departments suddenly face a queue of discovered flaws they may lack the staff to fix. Speed of detection has outrun speed of remediation.
What does this change for banks and their customers?
Monday's software glitch at JPMorgan Chase disrupted transactions for customers of other major banks — even though those banks' own systems were running normally, according to the Associated Press. The lesson: your uptime is only as good as your counterparties'. A payment rail you do not own can still take you down.
The pattern is not anecdotal. Third-party technology providers accounted for roughly two-thirds of publicly reported IT outages over the past nine years, according to a recent analysis from the Uptime Institute, a digital infrastructure advisory firm. Concentration, in other words, has been failing quietly for the better part of a decade.
The caveat behind every promise
The official story — old systems, upgrade strain, holiday volume — fits the facts on display. But it is worth noting what has not been disclosed: none of the affected companies has published detailed technical post-mortems yet, and the FAA's router explanation, while vivid, explains a trigger rather than the full failure chain. Why one router at one remote facility could cascade into a nationwide stop remains the question regulators will need to answer.
The broader takeaway for executives is unglamorous. Cutting-edge AI capability still runs on centralized, decades-old infrastructure underneath. Every promise of AI-driven efficiency carries a dependency on systems that were never designed for the load — and, as last week demonstrated, the bill for that dependency can arrive as 18,000 cancelled flights and frozen bank transactions in a single holiday season.
via fastcompany.com/rss (Source)
More from Sophie Lindqvist
Show full bio
Correspondent covering money, funding and market shifts around corporate strategy and chief-executive leadership at CEO Frontier.
479 articles
Adjacent sheets
- Four US disruptions in six days expose software patching strain
- Three U.S. Outages in Five Days Raise Unanswered DDoS Questions
- Microsoft outage and FAA halt test enterprise IT resilience
- Chase blames software bug as outages hit banking, aviation and Microsoft systems
- Four Days of Outages Exposed America's Aging Infrastructure Buttons