Artificial intelligence (AI) is expected to be a major ally of IT
security professionals in the fight against cyber crime in a few years.
“It will change the cyber security landscape,” said Singapore-based Christopher Church, digital forensic officer at the Interpol Global Cyber Complex for Innovation, at last week's Total Security Conference hosted by Computerworld Hong Kong.
He cited an AI system created by MIT that can prevent 85% of cyber attacks using input from human experts.
“The system will actually look for the attacks and will mark them; and then the user has to tell the system whether these are actual attacks or false positives. So it presents all the attacks discovered over the day, and then a human expert would say which ones are actual attacks. And once the attack data are put back into the system, it then reorganizes its logic, and then it becomes better at learning what it does,” Church explained.
This will ease the burden on IT security professionals who at present have to shift through reams of log data.
“AI will still need a helping hand, but human behavior is predictable 95% of the time. So things are getting better for cyber security,” Church said.
Cyber criminals are fast movers
Church noted that regardless of how “switched on” companies are in their cyber security posture, they can never outrun cyber criminals who are much quicker than the people policing corporate IT systems.
“The risks change every day. As soon as hackers find a new way of attacking you, it is circulated on the Internet. They let others know how they’ve done it,” Church pointed out.
He added: “We have seen a vulnerability found in one company being broadcast in a bulletin board, and within 20 hours of that posting, the company had been attacked over 2,000 times. And we actually watched it as it happened.”
According to Church, one cannot stop a determined hacker, so having a suite of security software and highly trained IT people are not necessarily the best protection against cyber crime.
“Software systems are engineered by humans and humans make mistakes. There will always be a vulnerability in a piece of software,” he said. “This is why you need to understand what and where your sensate data is held. Encrypt the data – make it a little difficult for them.”
However, security risks lurk on both sides of the wall. And Church – like many other security experts – pointed out insider threats exist within an organization. For one, the widespread use of social media, in particular, opens up a company to potential risks.
“The biggest security threat to your company is LinkedIn because hackers can use it to find out the company’s structure and who holds the payroll, among other things,” Church said.
And unfortunately, the systems users themselves are the weakest link in a company. “Older members of the staff may not be as technically savvy as their 20-year-old colleagues. So you can easily find that IT knowledge is pretty uneven within the organization.”
More than anything else, user education is the best way to combat cyber crime, Church said. “It is about having highly informed users.”
“It will change the cyber security landscape,” said Singapore-based Christopher Church, digital forensic officer at the Interpol Global Cyber Complex for Innovation, at last week's Total Security Conference hosted by Computerworld Hong Kong.
He cited an AI system created by MIT that can prevent 85% of cyber attacks using input from human experts.
“The system will actually look for the attacks and will mark them; and then the user has to tell the system whether these are actual attacks or false positives. So it presents all the attacks discovered over the day, and then a human expert would say which ones are actual attacks. And once the attack data are put back into the system, it then reorganizes its logic, and then it becomes better at learning what it does,” Church explained.
This will ease the burden on IT security professionals who at present have to shift through reams of log data.
“AI will still need a helping hand, but human behavior is predictable 95% of the time. So things are getting better for cyber security,” Church said.
Cyber criminals are fast movers
Church noted that regardless of how “switched on” companies are in their cyber security posture, they can never outrun cyber criminals who are much quicker than the people policing corporate IT systems.
“The risks change every day. As soon as hackers find a new way of attacking you, it is circulated on the Internet. They let others know how they’ve done it,” Church pointed out.
He added: “We have seen a vulnerability found in one company being broadcast in a bulletin board, and within 20 hours of that posting, the company had been attacked over 2,000 times. And we actually watched it as it happened.”
According to Church, one cannot stop a determined hacker, so having a suite of security software and highly trained IT people are not necessarily the best protection against cyber crime.
“Software systems are engineered by humans and humans make mistakes. There will always be a vulnerability in a piece of software,” he said. “This is why you need to understand what and where your sensate data is held. Encrypt the data – make it a little difficult for them.”
However, security risks lurk on both sides of the wall. And Church – like many other security experts – pointed out insider threats exist within an organization. For one, the widespread use of social media, in particular, opens up a company to potential risks.
“The biggest security threat to your company is LinkedIn because hackers can use it to find out the company’s structure and who holds the payroll, among other things,” Church said.
And unfortunately, the systems users themselves are the weakest link in a company. “Older members of the staff may not be as technically savvy as their 20-year-old colleagues. So you can easily find that IT knowledge is pretty uneven within the organization.”
More than anything else, user education is the best way to combat cyber crime, Church said. “It is about having highly informed users.”
Комментариев нет:
Отправить комментарий