COMMENTARY: Recent cyberattacks have disrupted refrigeration systems at European supermarkets and at military commissaries here in the U.S., forcing stores to throw out food.
At first glance, targeting refrigerators makes little sense. That's why I keep coming back to a different question: Is the spoiled food the point? Or, were those attacks a means to an end?
I've spent more than 30 years tracking advanced persistent threats (APTs), including during my time in military intelligence. I've seen patient adversaries spend months studying a target and preparing operations that may not happen for years. When I look at these incidents, I wonder whether someone could be testing which systems they can access, watching how operators respond, and gathering information they may use later.
[SC Media Perspectives columns are written by a trusted community of SC Media cybersecurity subject matter experts. Read more Perspectives here.]
Our adversaries play the long game.
APTs do not always go straight after the target they ultimately want. They gain access, learn how the environment works, and move through stages that can take months or years. The first intrusion often reprsents one step in a much longer program.
We've seen that kind of preparation before: In 2015, attackers maintained access for six months or more before disrupting Ukraine's power grid, leaving approximately 225,000 customers without electricity. The attackers also disrupted remote-control equipment and customer communications, forcing engineers to travel to substations to restore power manually. A decade later, attackers infiltrated a Polish combined heat and power plant, stole operational information and gained privileged access before trying to sabotage its systems. CERT Polska reported that security software blocked the destructive malware.
Refrigeration raises a similar question. There are only a handful of companies making many of these industrial systems. If a method works against one controller in Sweden, it may work against the same or similar equipment somewhere else. Testing in different countries could also make the larger pattern harder to recognize.
French authorities have now attributed the first wave of attacks in France to an Iranian military cyber unit. But we still don’t know whether the subsequent incidents across Europe are connected or the broader end goal of our adversaries.
What attackers learn
Attackers catalog which systems are vulnerable, how they fail, and in what order. They study how organizations and countries recover, what that recovery requires and where it starts to strain.
Gaining access tells an attacker which systems are vulnerable. Causing a disruption can reveal much more: how quickly operators detect the problem, how they recover, who's involved, whether specialists are needed and where resources become constrained.
Recovery can be intelligence, too. Which technicians do operators call? What backup equipment do they need? Where do they move inventory? What's the recovery path, and where does it start to strain? Each answer reveals another dependency that could attackers can disrupt.
Over time, a patient adversary could build a catalog of accessible equipment, successful intrusion methods, and recovery procedures across many organizations. The immediate damage from one incident is real, but it's not complete chaos. The long-game risk: someone could eventually string three or four of these events together and create a much greater impact.
Manageable outages are a national security problem
We can handle one refrigeration outage, that's fairly manageable. But several at once raises alarm bells. Stores may compete for the same technicians, generators, and backup facilities. There are only so many people and so much equipment available, so recovery slows as more organizations call for the same help.
That's the part I look at through my military intelligence background. There's established military logic behind this: study the systems, prepare access, and choose the moment to act. After cataloging these weaknesses and recovery dependencies, an adversary could sequence disruptions so that each makes the next harder to manage.
During a political or military crisis, that could force a government to divert scarce resources and leadership attention toward restoring daily life. The refrigerator is not the end of the story, it's quite possibly a means to an end.
My concern here: today’s incidents (including the wave of incidents over the past few days on airports, banks, and the NSA) could offer knowledge useful for a much larger operation tomorrow.
Protect critical infrastructure
We have known for years that critical infrastructure has vulnerabilities, from SCADA systems to other operational technology. It's not easy or trivial to protect those environments. Operational requirements and regulations complicate the work. They are not as simple as deploying a product, segmenting the environment, and declaring the job done.
Effective segmentation can limit the impact of an incident, but resilience also depends on what happens next. Organizations need to understand how their systems fail, what recovery depends on, and which shared resources become bottlenecks when several facilities need help at the same time. We need to design systems to fail gracefully, not catastrophically.
These attacks are bigger than any one vendor. Governments need to drive requirements for systems that support daily life, while operators, manufacturers and security companies make those protections work in real environments. We must make it a team effort.
The recent attacks are likely the paper cuts of a much larger problem. Attribution of one wave does not tell us whether every incident was connected or what comes next. But we do know that patient adversaries study access, impact, and recovery before they act. We should address these weaknesses while they are still paper cuts, not after someone decides to turn the dial.
Erik Bloch, vice president, trust and security, Illumio
SC Media Perspectives columns are written by a trusted community of SC Media cybersecurity subject matter experts. Each contribution has a goal of bringing a unique voice to important cybersecurity topics. Content strives to be of the highest quality, objective and non-commercial.