Palo Alto Networks CEO Nikesh Arora said AI helped hackers in most cyber events of the last 10 days. He wrote on X, "Most of the cyber events in the last 10 days had AI Assist." Hackers now use AI to automate attacks and target many organisations at once.
Autonomous machine learning frameworks are accelerating security breaches, enabling hackers to scale simultaneous multi-company attacks and overwhelming incident response teams. Enterprises are urged to abandon reactive tools and overhaul threat detection centers immediately.
Nikesh Arora, chief executive of cybersecurity company Palo Alto Networks, has warned that artificial intelligence (AI) is playing a role in most recent cyber incidents, as hackers increasingly use AI tools to automate attacks and target multiple organisations.
In a post on X, Arora wrote, "Most of the cyber events in the last 10 days had AI Assist." He did not identify the incidents. In a reply to another user, he said the cybersecurity industry had warned enterprises about the threat six months ago and that it was now materialising.
"The events of this week so far are not a coincidence. AI is making it easy for hackers to attack multiple customers," Arora wrote.
His comments come amid reports of AI-assisted cyberattacks. On 16 November, Google's Threat Intelligence Group reported identifying an AI framework called exploit_agent that could discover software vulnerabilities and develop and execute exploits without human intervention.
The framework analysed code changes introduced through security patches to identify potential weaknesses.
Separately, the FBI, in an advisory dated 13 November, warned that cybercriminals had been using an AI-powered penetration-testing framework called ARTEX to target organisations worldwide since at least August 2026. Some intrusions progressed from initial targeting to privileged access in under three hours.
Cybersecurity company Barracuda also disclosed a 44-hour intrusion that touched around 10 cloud accounts on 1 and 2 November. The attacker used a cloud access key, issued to a third-party AI vendor, that had never been revoked. Barracuda discovered the breach on 4 November and linked it to ARTEX.
Arora's advice to enterprises
Arora said incident responders were busy and that there was "no silver bullet" against the threat.
He advised companies to run a multi-model testing harness across their infrastructure, patch vulnerable code and explore virtual patching solutions for open-source software. Virtual patching provides temporary protection against known vulnerabilities without changing the underlying software code.
He also addressed how companies should respond when an attack succeeds. Their security operations centres (SOCs), the teams that monitor systems for threats, must be ready to act quickly.
"When it happens, because it will. Your SOC should be ready to detect and remediate in record time," he wrote.
Arora called for companies to overhaul their SOCs, recommending real-time data lakes, which are central stores of raw security data, along with proactive machine-learning (ML) detectors that spot unusual activity and AI-powered systems to correlate security information across their infrastructure.
"REVAMP your SOC to a real time data lake with proactive ML detectors across your data, with AI running correlation if you want a chance to come out ahead," he wrote.
He cautioned that reactive security measures spread across multiple data lakes were inadequate, adding, "Band-aid reactive solutions which work on multiple data lakes is not a plan."
