The market for cybersecurity keeps getting bigger, but the results are not getting better.
Over the past decade, the amount of money spent on cyber defense has exploded from less than $10 billion to roughly $70 billion, according to Symantec CTO Amit Mital. And all forecasts show spending continuing to skyrocket. But the litany of devastating hacks—at Target (TGT), Home Depot (HD), Sony, and the U.S. Department of Personnel Management, to name a handful—continues to pile up.
Cybersecurity is “basically broken,” said Mital during a panel discussion on Tuesday at Coins2Day’s Brainstorm Tech conference in Aspen, Colo.
The “bad guys” outnumber the “good guys” in the hacking world, said Mital, and they make more money, too. “You can’t out-hire them, and you can’t outspend them.”
Artificial Intelligence is one of the “few beacons of hope in this mess,” said Mital.
One crucial advantage that artificially intelligent defense systems would have is the ability to react instantly in real time. “You cannot have humans in the mix,” said Mital. By the time people recognize and take action to combat a hack, it’s often too late.
Mital argued that the spread of mobile technology and open source software is only making traditional cybersecurity more daunting, and the idea of a defendable “perimeter” is going away.
The question of how much cybersecurity authority to hand over to artificial intelligence systems, however, plays into a larger debate about how much trust humans have in machines, and what functions people will be willing to outsource to them.
Identifying functions where computers are better than humans is pretty clear today, but making decisions on where people want computers involved is a conversation that’s “not completely rational,” said Adam Nash, CEO of Wealthfront, an online money management service. Progress in embracing AI over the next five years, according to Nash, is going to come down to how much humans are willing to embrace it. That, he said, is a “harder question.”