As security pros scramble to defend against today’s threats as well as a new wave of attacks such as the WannaCry ransomware worm, emerging technologies and determined cybercriminals are poised to make cyber-defense even more complex.
Machine learning and the Internet of Things aren’t new terms, but they’re new to the workplace. As more businesses adopt advanced systems, they’ll find themselves vulnerable to a broader range of security threats. The geopolitical landscape will also drive security risk as nation-states target foreign adversaries without fear of punishment.
At last week’s Interop ITX conference in Las Vegas, security experts discussed the implications of current and future threats. Here are what they identified as the next big threats for enterprises:
The Internet of Things
IoT poses a tremendous security threat as users and devices become increasingly connected. The problem is, it’s so new that many people haven’t begun to worry about it.
“I was blown away to hear most people don’t think that’s really a thing yet,” said Dawn-Marie Hutchinson, executive director for the office of the CISO at Optiv, of the IoT during her Interop presentation on securing enterprise infrastructure.
The pressure to build for the IoT has already started. Companies rushing to cash in on the demand for connected products are churning out “smart” appliances, accessories, and other gadgets faster than they can secure them. Most people don’t bother to take precautions like changing default passwords, an oversight that’s leaving them vulnerable.
“We’re racing for innovation and disruption … but just because you can, doesn’t mean you should,” says Cheryl Biswas, a cybersecurity consultant for threat intelligence at KPMG. “We don’t need any of this stuff, but we are making it.”
The competition is driving an influx of Internet-connected devices built without basic security measures, and businesses and consumers will be at risk as flaws are discovered and exploited. The implications of poor IoT securitywere underscored with last year’s Mirai botnet attacks and the Persirai botnet discovered earlier this month.
FireEye CEO Kevin Mandia says attack strategies will continue to change and may evolve to the point where attackers will begin to exploit user trust. Human communication will “become the backbone for how we control devices,” he says.
“With wearable devices, all will be cloud-based, as we’ll all have physiology combined with technology, combined with the Internet,” Mandia continues. “Money, identity, everything will be part of the devices that you carry.”
Today’s criminals may find success with quick-and-easy attacks, but look for them to experiment with advanced techniques as machine learning and artificial intelligence as these technologies slowly pervade our everyday lives through Alexa, Amazon Echo, search results, and other instances.
The information that machine learning systems derive from rules, heuristics, signatures, and people will soar into the billions of pieces of information, according to Mandia. If the good guys are using it, we can bet the bad guys will use it, too.
Future attackers will exploit peoples’ reliance on machine learning and ignorance of how it generates results. Many users don’t take time to understand the processes behind these systems, instead trusting machine learning algorithms to find the shortest path to a result.
“Machine learning accepts what it’s given, and there’s no transparency into what went into a decision or model,” explains Anomali VP of product management Anthony Aragues.
He anticipates this will fuel the rise of “adversarial machine learning,” where attackers enter false information into systems to generate bad results. This could be used to disrupt services like facial recognition, as well as to misdirect people and conduct other attacks that abuse user trust in machine learning to target victims.
The risk of nation-state cyberattacks will also evolve amid today’s geopolitical landscape. We’re at the beginning of an age when “we’ll miss data breaches,” said Paul Kurtz, TruSTAR cofounder and CEO, during a talk on nation-state attacks at last week’s Interop conference.
“Everything we’ve seen so far … is nothing,” he continued, explaining how nation-state attacks will lead to more serious damage like incapacitated systems. These threats are so complex because adversaries share tools and strategies, and attacks are collaborative and automated.
Kurtz explained how adversaries like Russia and China are considered major powers because they have the influence to threaten victims “in an existential way.” Regional powers North Korea and Iran pose less of an immediate danger but are still a growing concern.
“We’re always on defense in cybersecurity,” FireEye’s Mandia said of the rise in nation-state threats during his Interop keynote. Foreign adversaries don’t need complex tactics; many exploit human trust. More than 90% of attacks FireEye investigates began with spearphishing.
Mandia noted FireEye is responding to more state-sponsored intrusions than financially motivated attacks. The rise in nation-state threats is partly due to a lack of punishment: there are no risks or repercussions to hackers, he continued.
“It’ll be about money, it’ll be about influence, it’ll be about espionage,” said Mandia of future attacks.
What to do about it
“With the IoT, it’s basically everyday things becoming digital,” said Daniel Miessler, director of advisory services at IOActive, during his Interop presentation on IoT security. “The problem is, businesses depend on those everyday things.”
He advised businesses adopting IoT take the time to conduct risk assessments before implementing products. Ask the questions: What data is being captured, and via what sensors? Where is it sent? How is it stored?
Those who use machine learning should not accept that a “magic algorithm” is producing good results, says Aragues. Take the time to understand how these systems work and interpret data so you can recognize and respond when something is suspicious.
On a consumer level, we need to emphasize basic security steps and explain the risks to users, says IBM Security’s global executive security advisor Diana Kelley. Cyberattacks will become more disruptive as we depend more on software and connectivity, she predicts.
“The key is making security accessible,” she says. “We need to help people help themselves.”
Kelly Sheridan is Associate Editor at Dark Reading. She started her career in business tech journalism at Insurance & Technology and most recently reported for InformationWeek, where she covered Microsoft and business IT. Sheridan earned her BA at Villanova University. View Full Bio