Skip to Content

Is Technology Failing? What Happened This Week and What to Do About It

What broke, why it happened, and how to keep your business running through the next outage
November 18, 2026 by
Is Technology Failing? What Happened This Week and What to Do About It
Casey Quinn

If it felt like technology fell apart over the past week, you're not imagining it. Email went down, flights stopped, banks slowed, and Apple's App Store went dark. Even the NSA had network trouble.


So, is technology failing? 


Federal officials say they've found no evidence that any of it was connected or caused by an attack. What these outages do share is less dramatic and more useful to know: nearly every one started with a change, like an update, a patch, or new equipment.


Here's what happened, what caused it, and what you can do so the next one doesn't stop your business.


Date

What went down

What people couldn't do

Stated cause

Thu, Nov 12

Microsoft 365 

(Outlook, Teams, Copilot)

Send email, manage calendars, or hold online meetings for about a day

A software bug from a July update, triggered by a traffic surge and reduced server capacity

Fri, Nov 13

FAA air traffic network

Fly: departures nationwide were held from about 3:30 to 8:15 p.m. ET, and about 12,462 flights were canceled

Routers at two remote sites flooded the FAA's network with data; the exact cause is still under investigation

Fri, Nov 13

Amtrak booking

Book or change train tickets online, right as stranded flyers looked for another way home

Amtrak pointed to a systemwide update

Fri, Nov 13

Workday

Clock in, request time off, or finish open enrollment at companies using the platform

Reported to have surfaced during testing of backup systems

Mon, Nov 16

Chase and other banks

Complete some transactions on time; Wells Fargo online banking was unavailable that afternoon

Chase: a software release

Mon, Nov 16

United Airlines

Depart on time: 418 United flights were delayed that afternoon

Not stated; pilots pointed to the system that calculates takeoff weight

Tue, Nov 17

NSA internal networks

Use internal networks, including a classified communications network

Scheduled maintenance; patched systems didn't come back up as expected

Tue, Nov 17

Apple services

Use the App Store, Apple Music, and AppleCare; iCloud sign in also had issues

Apple pointed to an internal update

Marriott also reported a technical issue that blocked sign ins on its website and Bonvoy app, and said it found no sign of a cybersecurity problem.

Was it an attack? 


It's a fair question. When this many big systems break in the same week, it's natural to wonder whether someone is behind it.


So far, the answer is no. None of the companies or agencies involved has blamed a cyberattack, and federal officials said they've found no evidence tying the outages to each other or to foreign hackers or governments.


The Director of National Intelligence put it in plain terms. Institutions were patching security flaws and taking systems offline, and things failed when one system tried to hand work off to another.



Where AI fits in.

AI came up a lot this week, so here's what has actually been said.


Virginia Sen. Mark Warner, vice chairman of the Senate Intelligence Committee, said AI is speeding up how fast software flaws get found. That puts pressure on organizations to patch quickly, which strains their systems. He said he was not suggesting a cyberattack.


There's a recent example. In October, researchers working with AI tools helped uncover two critical flaws in PHP, the programming language behind nearly 70% of websites whose server language is known. The FBI and NSA advised taking systems offline if they couldn't be updated quickly, and this week's NSA outage came while it was patching a PHP vulnerability.


University researchers interviewed this week added two points. One said companies using AI are finding bugs faster than people can and rushing the fixes. Another said that as more software is written by AI, people need to be skeptical of it rather than trust it blindly.



If your business uses AI, watch for this:


  • Check AI's work before it goes live. Code, scripts, formulas, and automations written by AI should be reviewed and tested by a person before they touch live systems or customer data.
  • Limit what AI tools can change on their own. A tool with permission to edit live systems can make a bad change as fast as a good one.
  • Know what your AI tools depend on. During the Microsoft outage, AI tools from OpenAI and Anthropic had trouble with their Microsoft connections. If an AI tool runs on top of your email or files, it goes down when they do.
  • Plan for more updates, not fewer. If flaws are found faster, patches arrive faster. Skipping them isn't the answer: the PHP flaws were serious enough that federal agencies said to go offline rather than stay unpatched.



What actually caused it?

Look at the cause column again and a pattern shows up. Almost every outage followed a change someone made on purpose:


  • Microsoft: a software bug from a July update, triggered when traffic surged and some server capacity was already offline.
  • FAA: routers at two remote sites flooded the network with data, including a new one being installed that day. Investigators haven't yet determined whether it was an installation error or a faulty router. The FAA has switched off the function involved on its new routers and plans to rethink scheduling, since this work happened in the middle of the day.
  • Chase: a software release.
  • Apple: an internal update.
  • NSA: security patches, after which some systems didn't come back up.
  • Amtrak: a systemwide update.

These are the same kinds of changes every business makes. When one goes wrong at a company millions of people rely on, everyone who depends on it is affected at the same time.


A cybersecurity researcher at the University of Alabama at Birmingham told WBRC that outages like these are common for large systems. This week stood out because they came back-to-back.


What the week looked 

like for our clients.

Our phones were busier than usual all week. Most of the calls were about outages and servers that were down because of these incidents.


When Microsoft went down on Thursday, our technicians recognized it as an outage before Microsoft announced it publicly. That let us give clients a straight answer early. We worked through the risk with anyone who had lost access or wasn't getting email, including moving staff onto workarounds where needed.


Managed clients got priority on all of it. When everyone is calling at once, they're first in line.

What to do about it.

Our Founder & CEO's take.

Christian Argie founded Top Notch Computers in 1999. When CBS19 asked him about the Microsoft outage, 

he said most businesses don't have a plan for an outage, and that a good one avoids relying entirely on one vendor. "It could have been way worse," he said. "There could have been data loss."


Asked whether technology is failing, here's how he sees it:


"Technology isn't failing. It's changing faster than it ever has, and every change is a chance for something to break. The businesses that got through this week weren't the ones with perfect technology. They were the ones with a backup for the things they can't control."

If you run a Business...


Know what you can't run without

List the outside services your business depends on, like email, payroll, banking, phones, and your scheduling software. For each one, ask what you'd do if it went down for a day.

Back up your Microsoft 365 data outside Microsoft

A separate backup means an outage doesn't turn into lost data.

Test your backups, don't just have them

An expert told Inc. this week that backups that look fine on paper often 

fail when a real outage hits.

Have a backup way to communicate

Decide now how your team will reach each other and your 

customers if email and Teams go down.

Treat updates as the risk they are

Schedule them outside business hours, test them when you can, and have a way to undo them. The FAA is rethinking midday maintenance for exactly this reason.

Know who you'll call

Know how fast they'll respond when every other business is calling too.

For everyone...


Carry a second way to pay

A University of Alabama at Birmingham researcher advised against relying only on mobile payments like Apple Pay and suggested keeping a physical card or some cash.

Watch for scams

Scammers use the confusion around outages. Confirm anything about your accounts directly with your bank or provider and never share personal information with 

someone who reached out to you.

Lock down your accounts

Turn on multi-factor authentication for banking and other sensitive accounts 

and use cellular data instead of public Wi-Fi for banking apps.

Check the source

If a service seems down, look at its official status page or verified social accounts 

before assuming the problem is on your end.

How Top Notch helps

No IT provider can fix Microsoft's servers or the FAA's network. What we can do is keep someone else's bad day from becoming lost data or a lost week at your business. 


For managed clients, that means:


  • Microsoft 365 backup, so a problem at Microsoft doesn't mean losing your data.
  • Watching the services you depend on and telling you when one breaks, like we did when Microsoft went down.
  • Backup and recovery for data kept in your office.
  • Workarounds for your staff when a service is down.
  • Priority response: 24/7 coverage, acknowledgement and active work within 2 hours, and a technician on site within 4 hours if needed.

Since so many of this week's outages started with an update, here's how we handle updates for managed clients:


  • Installed after hours, so a bad update doesn't take you down in the middle of the workday.
  • Rolled back if an update causes trouble.
  • Monitored afterward, so problems get caught and fixed quickly.


Where does your business stand?

Not sure how your business would hold up if one of these outages hit you directly? Our free Technology Risk Assessment takes a few minutes and scores how exposed you are when a vendor, a network, or a system goes down.


Take it at topnotchcomputers.com/risk-assessment, and our team will follow up to talk through your results.


Top Notch has provided managed IT services in Virginia since 1999, with offices in Charlottesville, Richmond, and Falls Church. We've served more than 5,000 clients, we keep 97% of our clients year over year, and our current clients include banks, credit unions, and accounting firms.


Sources



Share this post