No. 457 Nov 18, 2026

Attack or Outage?

In October, attackers pushed the freezers in hundreds of French supermarkets to 122°F without ever setting foot in a store. Last week, flights across the country were grounded, bank customers couldn’t reach their money, and much of the internet decided it was all connected. The freezers were a confirmed attack. Most of last week, as far as anyone can tell, was not.

This issue is about telling the difference: what infrastructure is, what a coordinated attack would actually look like, and what’s in our hands when something big goes down.

Digitally Literate: A weekly newsletter and connected notebook for anyone trying to make sense of AI and the digital world, and keep some power over it.


Reading the signs

Infrastructure is the stuff you only notice when it breaks: the water that comes out of your tap, the power that keeps the lights on, the grocery store freezer that keeps your ice cream frozen overnight, the system that sends an ambulance when someone calls 911.

We tend to picture pipes, wires, and concrete, but over the last few decades much of this infrastructure has come to depend on computers. Pumps, valves, thermostats, and dispatch systems now use small networked devices, and many can be monitored or controlled remotely, so nobody has to stand beside a machine to change how it works. That convenience also creates an opening.

Cybersecurity is the work of protecting computers, networks, and connected systems from interference, theft, or damage. A cyberattack is a deliberate attempt to break into or disrupt those systems. Attackers might steal passwords or medical records, or use ransomware, which locks an organization out of its own systems until it pays. An attack doesn’t have to steal a single file to hurt people, though.

Sometimes an attack is part of something larger. A coordinated cyberattack is a campaign in which one group targets multiple targets simultaneously or sequentially, often causing more disruption than any single attack could. Most of us wouldn’t recognize one in its early days. Outages happen all the time, and from your phone, a bank that’s been hacked and a bank with a software bug look exactly the same.

I keep thinking about what happens when a hurricane heads toward town. Most of us watch the sky and the news. The people who’ve been through a few watch for specific things, like the forecast cone narrowing over their county or officials ordering evacuations instead of suggesting them, and they know which signs mean board up and which mean wait and see.

Large cyberattacks have signs like that too:

Until signs like those show up, the most useful thing most of us can do during an outage is wait for the second explanation, and notice who gives it.

Someone is turning up the heat

Overnight on October 11, refrigerators in hundreds of French supermarkets started heating up. Attackers had reached the systems that control them remotely, pushing temperatures as high as 122°F (50°C) and switching freezers into defrost mode. Food spoiled, stores threw away their stock, and some locations had fires.

On November 5, five U.S. federal agencies, including the FBI and the Cybersecurity and Infrastructure Security Agency (CISA), issued a warning about attacks on industrial refrigeration and cold-chain monitoring systems. Nobody had to break into a grocery store or sabotage a freezer in person. Someone reached a computer connected to the cooling equipment and changed its settings.

These systems aren’t limited to supermarkets. Hospitals, food distributors, pharmaceutical companies, and transportation networks rely on similar equipment to keep supplies at safe temperatures, so a change on a screen can make food unsafe or medicine unusable. The same concern applies to water treatment plants and electrical systems: when computers control physical equipment, someone who gains access to the computer may be able to interfere with the physical world.

A week after that warning, many of us felt like everything was breaking at once. On Thursday, November 12, Microsoft 365 went down worldwide, taking Outlook and Teams with it. The next afternoon, the Federal Aviation Administration (FAA) stopped departures across the country for about four hours, only the second nationwide ground stop since 9/11. On Monday, United Airlines halted departures, and customers at Chase, Bank of America, Wells Fargo, and other banks couldn’t access their accounts. That same day, researchers found malicious code running on CNN, the BBC, and dozens of other sites through scripts hosted by the marketing company Optimizely.

Online, those events were quickly stitched into a single story about AI-powered attacks and companies covering them up, and the theories spread faster than the explanations did. We don’t know yet whether any of them are connected. So far, most of the organizations involved have pointed to ordinary causes, and the one confirmed attack, at Optimizely, hit websites rather than airlines or banks.

What’s in our hands

It’s easy to read stories like these and decide that cybersecurity is either too technical to understand or too big to do anything about. More of it is in our hands than it seems, and most of it is habit rather than expertise.

A few habits cover most of the everyday risk:

CISA’s Secure Our World site walks through each of these, from creating strong passwords to turning on two-step login.

Getting ready for a large outage turns out to be mostly ordinary disaster preparation, the same kind you’d do before that hurricane. Picture a few days when the internet, card payments, or your phone don’t work reliably, and think about what you’d need:

When something big does happen, expect bad information. Rumors move faster than official channels, so wait for confirmation from your local emergency management office, your utility, or your bank. Scams follow the news closely, and fake “fix your account” emails, texts, and calls tend to show up within hours. No legitimate company will ask for your password to restore service, and you can report messages like that to the FTC. Resist the urge to panic-buy or pull your money out, since most systems come back within hours or days. Once things are working again, check your accounts for any odd charges or logins, and change your passwords if a company you use has been breached. Have I Been Pwned will tell you whether your email address has turned up in a known breach.

There’s a limit to individual responsibility, though. You can’t update the software controlling your grocery store’s freezers, secure a hospital’s equipment, or audit every company that handles what you type into an AI feature. Those responsibilities belong to the organizations that build, buy, run, and regulate these systems.

The Understory

In the middle of the 1800s, the most advanced technology in American life was the steam boiler. It drove the riverboats that carried people and cargo up and down the Mississippi, powered factories and mills, and heated buildings in growing cities. It was also dangerous in a way you couldn’t see. A boiler held water under enormous pressure, and when one was built poorly, run too hot, or simply worn out, it could explode and tear apart the boat or building around it.

On the rivers, explosions were common enough to make the news. In the eight months before Congress acted in 1852, seven steamboat boiler explosions killed more than 700 people. The Steamboat Act of 1852 set standards for how boilers had to be built, required safety valves and limits on operating pressure, licensed the people running them, and created a corps of federal inspectors under the Treasury Department to check that boats met the rules. Legal scholars describe it, along with an earlier law from 1838, as among the first federal regulations of a private industry.

Inspection didn’t end the danger on its own. In 1865, the inspected and certified steamboat Sultana exploded on the Mississippi while carrying about six times as many people as it was approved for, and it remains the deadliest maritime disaster in U.S. history.

On land, part of the answer came from a group of engineers in Hartford, Connecticut. In 1866, they founded the Hartford Steam Boiler Inspection and Insurance Company on a simple idea: inspection made boilers safer, insurance gave owners a reason to get inspected, and a company paying out claims had every reason to inspect carefully. Its standards for how boilers should be designed, built, and maintained became the ones the industry followed.

Today most of us never think about the boiler in the basement of a school or an office building. Somebody decided it had to be checked, and the checking became ordinary enough to disappear.

The answer for boilers wasn’t teaching everyone to be a boiler engineer. It was deciding that someone had to inspect them, which is the question we’re facing now with the computers behind our freezers and water pumps.

See you next week.


Follow the ideas

Ian O’Byrne is a literacy researcher and professor exploring power, agency, and who gets to decide how technology shapes our lives.