Hackers Leverage AI to Power Cyberattacks

Thanks to artificial intelligence, hackers are on the verge of launching fully automated cyberattacks that can move faster, smarter, and more personally than anything seen before — potentially crippling critical infrastructure before anyone even realizes something is wrong.

These AI-powered attacks could halt production at factories, knock hospitals offline, or even disrupt power grids. As generative AI becomes more advanced, experts warn it’s giving cybercriminals unprecedented capabilities to boost their skill sets and automate complex stages of the attack chain.

Both OpenAI and Anthropic have already detected evidence of nation-state adversaries and cybercriminals using AI models to write malicious code and research new attack methods. Sandra Joyce, head of Google’s Threat Intelligence Group, said her team has also observed hackers attempting to exploit legitimate AI-driven hacking tools.

“Nation-state hackers are going to build tools to automate everything — from spotting vulnerabilities to launching customized attacks on company networks,” said Phil Venables, partner at Ballistic Ventures and former chief security officer at Google Cloud.
“It’s definitely going to come. The only question is: Is it three months? Is it six months? Is it 12 months?”

A recent Microsoft report underscored the danger: AI-generated phishing emails achieved a 54% click-through rate — more than four times the success rate of traditional phishing attempts.

Cybersecurity firm Deep Instinct found that 50% of professionals working in critical infrastructure reported facing an AI-powered attack within the last year. The survey also showed hackers are increasingly focusing on critical infrastructure and financial services, where disruptions could have far-reaching consequences.

China, Russia, Iran, and North Korea are already embracing AI to enhance espionage and cyber operations, security executives told Axios.

Chinese hackers are using AI “as a side saddle” to augment influence campaigns, according to Joyce. Meanwhile, Russian government hackers have experimented with AI-powered malware in their ongoing attacks on Ukrainian targets.

The emergence of apps like OpenAI’s Sora, capable of generating realistic videos, adds another layer of risk. Such tools could be used to create convincing deepfakes — for example, a celebrity promoting a fake investment or a video falsely depicting a child in danger — to deceive or manipulate the public.

Despite the growing threat, cybersecurity leaders say there’s reason for hope. Many believe the same AI tools used by hackers can also be wielded for defense — enabling faster detection, smarter responses, and stronger systems overall.

Defenders envision AI that can instantly sift through hundreds of threat alerts, identify genuine risks, and neutralize them in seconds. AI is also showing promise in writing more secure, flaw-free code.

According to Deep Instinct, more than 80% of major companies are already deploying AI to strengthen their cybersecurity posture.

Wendi Whitmore, chief security intelligence officer at Palo Alto Networks, said automation has already made a measurable impact. In one case, her team helped a major transportation manufacturer cut its attack response time from three weeks to just 19 minutes.
“We’ve got so many more layers of defense,” Whitmore said. “I can talk myself into being completely optimistic about AI.”

Experts like Jen Easterly, former head of the U.S. Cybersecurity and Infrastructure Security Agency (CISA), predict that autonomous AI-driven cybersecurity systems will soon be able to identify vulnerabilities beyond human capability — spotting intrusions before they occur and deploying countermeasures in milliseconds.

If those systems can also learn and adapt from each response, the balance of power in cyberspace could shift once again — this time, in favor of the defenders.

About the Author

Wayne Creed

Administrator

View All Posts