AI called upon as cyber crime surpasses human capability

Humans can no longer keep up with the evolving cyber crime industry and must now use AI models to help fight off bad threat actors.

BlackBerry Asia Pacific and Japan director of engineering Jonathan Jackson.

The cyber crime industry has already evolved beyond human capability and is increasingly seeing the adoption of artificial intelligence models to help stop bad threat actors from stealing sensitive data.

Humans alone are no longer capable of stopping large-scale attacks at the necessary speeds to prevent massive data breaches, according to BlackBerry’s cyber security team.

The Canadian company, once known for mobile phones, has pivoted away from handheld devices and into the cyber security field where it has landed a number of contracts with governments and large enterprises including in Australia.

It has around 3500 employees globally with three local offices in Canberra, Melbourne and Sydney, with staff in the fields of sales engineering, HR, finance and public relations.

Speaking to The Australian during a visit to the country, BlackBerry Cyber Security president John Giamatteo said most people weren’t aware of the company’s pivot and a number of government and secret contracts limited its ability to advertise and share who its clients were.

“When governments represent a big part of what we do, sometimes they want us to be a little bit quiet and that handcuffs us a little bit because we’ve got to respect their wishes,” he said.

Australia is in the world’s top five most cyber-attacked countries, and the company is increasingly looking to establish ground here.

BlackBerry Cyber Security president John Giamatteo.

Jonathan Jackson, BlackBerry’s Asia Pacific and Japan director of engineering, said the adoption of AI to fight cyber crime has been a long time coming. “We’ve been anticipating AI being used by cyber attackers for a very long time,” he said.

“Recently we’re starting to see evidence on underground hacking forums that cyber criminals, both nation-state as well as crimeware, are looking to use things like large language models to be able to improve their cyber attack capability.”

AI has become better known in recent months after the emergence of ChatGPT.

“Everybody’s now an AI expert because of ChatGPT but the reality is that not all AI is equal. In cyber defence technologies and capabilities, we’ve got to make sure that we are looking to continue to be able to fight AI,” Mr Jackson said.

“Humans are no longer able to keep up with the evolving cyber threat. The scale and the tactics and the ability of cyber criminals to be able to pivot their attacks on an expanding threat surface means that it’s not realistic for us to use humans alone to try and prevent cyber attacks.”

BlackBerry’s use of AI to defend cyber crime includes a threat detection system which uses math models to determine whether online activity is good or bad within milliseconds. The engine, called Cylance AI, is in its seventh iteration.

“We’ve got a very good understanding of the way malware code is written as well as the way that cyber attackers will try to bypass defence techniques,” he said. “Whether it’s the use of stolen credentials or anonymous behaviour on a network, we can use machine learning and AI to detect that at machine speed and block that attack early on in the cyber kill chain.”

BlackBerry Asia Pacific and Japan director of engineering Jonathan Jackson.

Originally published as AI called upon as cyber crime surpasses human capability