We now have the first evidence that the emerging Internet of Things is also the Internet of Things That Can Deliver Spam. A security firm has uncovered a global cyberattack that harnessed connected household devices, including a refrigerator.
Proofpoint Inc., based in Sunnyvale, California, said the attack utilized 100,000 consumer devices, employing them as other attacks have used captured computers -- to secretly deliver spamming e-mails numbering in the hundreds of thousands. The attack took place recently, between Dec. 23 and Jan. 6, and included in its botnet at least one smart refrigerator, as well as home-networking routers, connected multimedia centers and smart TVs.
The security company told news media that the attack "may be the first proven Internet of Things-based cyberattack involving conventional household 'smart' appliances." The attack, Proofpoint said, was sent in bursts of 100,000 e-mails three times daily, directed at companies and individuals around the globe, and over one-quarter of the spam was sent by compromised, non-computing devices.
No single IP address was used to send more than 10 e-mails, which made the attack's origin more difficult to locate, and the commandeering of the devices involved the relatively simple tasks of taking advantage of misconfigurations and default passwords.
David Knight, general manager of the Information Security Division at Proofpoint, said in a statement that devices in the Internet of Things "are poorly protected at best and consumers have virtually no way to detect or fix infections when they do occur." Businesses, he said, may find distributed attacks "increasing as more and more of these devices come on-line and attackers find additional ways to exploit them."
While not surprising, this new turn in malware raises a variety of questions about the addition of intelligence and connectivity to appliances and other non-computing products. If they can become "thingbots" and commandeered for nefarious purposes, will they also need the kind of anti-virus software, updates and continual vigilance that users are now required to do for their computers and mobile devices?
Application Control, ISP Vigilance
And how much additional effort and expense will it take to build and monitor a security fence around intelligent TVs, refrigerators, stoves, lighting devices in homes, connected coffee pots, smart thermostats and the like? According to some estimates, the Internet of Things already includes more than 2 billion devices, and industry research firm IDC has predicted there will be more than 200 billion connected things by 2020.
Peter Firstbrook, an analyst with Gartner, told us that "manufacturers have to get smart" about making all of these devices attack-proof, primarily by utilizing "application control, not anti-virus software," so that a connected device is built to run only specified applications and accept updates only from a single source, such as Westinghouse.
But some devices have become, essentially, computers able to run a large variety of applications, such as smart cars or connected TVs. Protection for those devices in particular, Firstbrook said, will need to happen at the network level, through a third party such as Zscaler or through an ISP, providing a continual scanning of all incoming and outgoing traffic.