By Matt AgoristThe company is working to develop an automated system that can detect if people are infected by malicious software and identify their actions before they are malicious enough to be exploited.
“The system is based on the premise that it’s better to know the person has been infected before they take action,” said Peter Smith, vice president of global intelligence for Verizon, at a recent event on cyberthreats.
“If we can do that, then we can detect it much earlier.”
Smith’s company, the Global Engagement Lab at the University of Virginia, has been studying how to build the system and is working with Microsoft Research to build it, as well as with other organizations.
Smith said that while he thinks it’s a good idea to monitor users, “I don’t think it’s really practical” for everyday users.
What’s more, Smith said, it’s important to know if someone is infected before taking any action to avoid inadvertently infecting others.
“We don’t want people getting infected and then thinking, ‘Oh well, I should stop, because that’s what we were told to do,'” he said.
“We want to be able to say, ‘You know what, you did it, you know what?
You could have stopped this.'”
He also said the system could be used to monitor and identify users who are actively spreading malware, but said it could also be used for tracking and tracking down users who aren’t actively using malware to spread it.
While Smith emphasized that this would be a pilot project, the idea of tracking and monitoring activity to determine if a user is infected was first introduced at a cybersecurity conference earlier this year, and it’s been widely adopted by the industry.
Microsoft has also been testing the technology, which it calls a “threat attribution” system.
The system uses a computer-generated image of a person and their actions to determine what actions the person might have taken in the past to help prevent future attacks.
The system would also analyze the activity of the person in question to identify what kinds of actions are likely to lead to further infection and then determine how best to respond to the person’s actions, Smith added.
For example, if someone who is actively spreading malicious software has uploaded a malicious document to a social media account, a system would determine what kinds that person might be likely to post to encourage others to follow suit.
“This will be a really valuable tool for us to be really careful with, because it’s one of the key technologies that will enable us to take on the threat and to stop it,” Smith said.
At the same time, the system is designed to be highly resistant to being hacked.
It would also only work if the user had permission to do so.
Smith said the company is not aware of any security breaches in its systems, but added that it does not have any malware in the system that could be exploited by a malicious party.
“That is our biggest fear, that people will try to exploit this to exploit their own data,” he said, adding that the system has been built with robust security protections.
“If you’re the company, if you’re running this, I would ask that you do the right thing,” he added.
“I think it will be an important tool in the future for all of us to do our part in protecting our data.”