This is not a cyberattack. It is the bill for years of accumulated vulnerabilities finally coming due.

From November 12 until today, the world has witnessed a series of overlapping technological disruptions:

▪️ Microsoft 365 services went down worldwide, including Outlook, Teams, and Copilot.

▪️ The US Federal Aviation Administration (FAA) grounded all flights across the United States.

▪️ Chase Bank and United Airlines experienced service disruptions.

▪️ Amazon Web Services (AWS) encountered technical problems in its us-east-1 region.

▪️ Apple services, including iCloud and Apple Pay, went offline.

▪️ Reports of Gmail outages surged dramatically.

▪️ Even the US National Security Agency’s (NSA) classified network, known as NSANet, experienced disruptions during the same week.

Why Is All This Happening?

Seven outages, seven different official explanations, but one underlying disease: excessive centralization.

Microsoft itself confirmed details revealing the scale of the problem.

A defect had remained hidden in its networking gateways since July. It later coincided with maintenance that reduced capacity by 14%, followed by a change in network traffic that caused more than 80% of gateway servers to become unresponsive.

It happened simultaneously, on a global scale.

The secondary failover system, which was supposed to take over automatically when the primary system failed, could not intervene because the abstraction layer designed to isolate different regions had also failed.

Even worse, the recovery process triggered another system collapse due to a massive simultaneous surge in requests, known as a Request Stampede.

Engineers ultimately had to manually disable automated traffic routing to give the servers enough breathing room to recover.

🔻 Why Now?

Because hidden software defects are accumulating faster than our ability to detect them.

The defect that emerged in July remained embedded in the system for four months because automated rollback mechanisms cannot detect a change that has gradually become part of the system’s normal operating baseline.

After that, a single maintenance window is enough to ignite the fuse.

🔻 Why Is There No Real Backup Plan?

Because what we now call a

backup plan

has become little more than additional software running on the same underlying infrastructure.

Failover itself has become a software service.

Redundancy exists by design, but it is not genuinely independent.

The same code, the same team, the same cloud infrastructure, and the same cost model.

When the primary gateway fails, the

“other gateway”

fails with it.

A genuine backup plan should provide an entirely independent secondary route, kept as simple as possible:

Paper-based systems, satellite communications, or out-of-band communication channels operating separately from primary networks.

But we gradually dismantled these alternatives in the name of

“greater efficiency and lower costs.”