When hackers target grocery stores
A cybersecurity expert explains what recent attacks on supermarkets reveal about how digital threats can reach into everyday life—and how AI-assisted attacks could escalate these risks
Tufts University
Recent cyberattacks targeting commercial refrigeration systems and smart temperature controllers in supermarkets across Europe remotely raised their temperatures—in some cases to as high as 122° Fahrenheit. The attacks spoiled large amounts of fresh and frozen foods, sparked small fires, and cost millions of euros in estimated product and equipment losses.
The attacks unfolded in waves, explained Shaanan Cohney, an assistant professor of computer science at Tufts University who researches the intersection of cybersecurity, privacy, law, and public policy.
“First, in mid-October, 700 supermarket locations in France were targeted,” he said. “After a brief pause, another 100 locations across Europe were hit in early November. French authorities took the unusual step of publicly attributing the initial attacks to Iranian state-sponsored hackers.”
Similar freezer malfunctions that spoiled food at commissaries on several U.S. military bases in August now are also under investigation as suspected cyberattacks. The Cybersecurity and Infrastructure Security Agency, FBI, and other U.S. federal agencies on Nov. 5 issued a joint security advisory, given the global threat to keeping food, medicine, and other products at safe temperatures.
Are these attacks targeting new or unusual targets?
There is a long history of attacks on operational technology systems, from traffic signals and subway fare card systems to the refrigeration systems targeted in this latest wave of attacks. What’s new is not that these devices are being attacked, but the scale and apparent ease of the attacks, potentially by a foreign nation.
Do appliances such as refrigerators being connected to the “internet of things” create new risks?
Connecting devices to the internet is nothing new. But what’s different here is that a state-sponsored attacker appears to have coordinated attacks against critical infrastructure that is highly visible to the public. There have been attacks targeting water systems and electric utilities, but targeting supermarkets brings the threat home to consumers in a much more immediate, everyday way.
Was the motive to show the public their country could be attacked?
It’s hard to know the motivation behind these attacks. But we’re increasingly seeing countries make their cyberattacks visible to the public, seemingly to influence public opinion and political outcomes.
We’ve seen Iran leverage social media and public opinion in a way that suggests they’re not only shaping geopolitics throughout the recent conflict. These supermarket attacks likewise may be both an attempt to disrupt ordinary life but also a way of trying to influence public opinion.
How serious is this type of threat to the security of our food supply chain?
These particular attacks aren’t going to bring down our food supply chain. But they highlight a broader risk as more of the systems we depend on become connected to the internet.
It’s increasingly difficult to just walk into a store and make an ordinary purchase without interacting with a networked system. And as our grocery stores, transportation systems, and healthcare become more reliant on digital technology, the risks grow, too.
Are refrigeration systems particularly vulnerable to hacking?
Yes. Many operational technology systems are built from components supplied by a large number of manufacturers. Each digital component runs its own highly tailored software or firmware—the software embedded in devices to control how they operate. Because many of these manufacturers may face few security standards and are not regularly patched when flaws are found, the components are often not particularly secure.
As a researcher, it’s very uncommon to examine one of these devices and find a clean bill of health. Often, you find not just one security flaw, but many. That leaves these systems vulnerable to state-sponsored attackers, who may need to find only one weakness to bring an important system down.
Were the grocery store attacks carried out with the help of AI?
It’s unclear if or how much AI contributed to these particular attacks. But AI is making cyberattacks more accessible to a wider range of bad actors. Whether or not these attacks were AI-powered, they offer a glimpse of what may be ahead.
AI is changing the landscape. Tasks that might once have taken a well-trained attacker weeks to accomplish can now be done by someone with far less expertise, with help from AI. In some cases, AI can even identify and link multiple security vulnerabilities to carry out an attack.
These capabilities are already shaping cybersecurity policy in the U.S. and internationally, which so far are often focused on trying to add safeguards to slow the development and spread of these models in countries that could pose a security threat.
However, recent headlines suggest these efforts have been less effective than officials would like—and restricting access to chips alone won’t provide a long-term solution to the cybersecurity challenges posed by AI. Government and industry are also trying other measure like using AI for cyber-defense and trying to build systems that are secure from the get-go, but there is no single policy fix.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.