Cybersecurity entering AI-vs-AI era as attackers, defenders deploy autonomous systems: WEF report

Cybersecurity is entering an “AI versus AI” era, with firms more and more deploying autonomous synthetic intelligence programs to research, detect and reply to cyberattacks at machine pace as hackers themselves quickly undertake AI-driven assault strategies, based on a brand new World Financial Discussion board (WEF) white paper launched in collaboration with KPMG.The report, titled “Empowering Defenders: AI for Cybersecurity,” stated attackers are more and more utilizing AI “to extend the pace, scale and class of threats,” forcing cybersecurity programs to evolve past conventional human-led defences.

“Adversaries are more and more working at machine pace, utilizing AI to conduct reconnaissance of targets and vulnerabilities, generate malware, exploit code, evade detection and launch assaults at scale,” the report stated. The WEF report famous that what earlier required “weeks of effort can now be executed in minutes,” decreasing technical limitations for cybercriminals and considerably rising the size of assaults. To counter this shift, organizations are more and more deploying AI-driven programs able to autonomously analysing threats, investigating suspicious exercise and recommending or triggering response actions in actual time.

Highlighting real-world deployments, the report cited IBM’s “ATOM” system, which autonomously investigates, enriches and scores cybersecurity alerts utilizing agentic AI. In line with the report, the system now handles “about 95 per cent of each day investigations” whereas automating greater than “850 analyst hours per thirty days.”

The report additionally pointed to Allianz’s “hypothesis-based AI evaluation system,” which dynamically retrieves and analyses forensic information throughout investigations as an alternative of accumulating all endpoint information centrally. The system was developed to deal with the rising problem of analysing huge volumes of cybersecurity information generated throughout enterprise programs.

In the meantime, Google has deployed AI brokers such as “Massive Sleep” and “CodeMender” to determine unknown software program vulnerabilities and mechanically generate safety patches. The report stated CodeMender has already patched “greater than 100 vital safety points.” The report added that AI adoption in cybersecurity is accelerating quickly throughout industries, with 77 per cent of organisations already utilizing AI in cybersecurity features.

In line with the report, organizations extensively utilizing AI in safety operations diminished common breach prices by “$1.9 million” and shortened breach timelines by roughly “80 days.” The WEF report recognized AI as “probably the most vital driver of change in cybersecurity,” citing findings from its World Cybersecurity Outlook 2026 survey, the place 94 per cent of respondents highlighted AI’s rising impression on cyber defence.

Nonetheless, the report cautioned towards extreme dependence on AI-driven programs, warning that “heavy reliance on AI can undermine cyber resilience.” “Extreme belief in automated selections creates a false sense of safety and over time erodes the experience wanted to intervene when programs fail,” the report stated. The report burdened that whereas AI is changing into central to cybersecurity operations, “human judgement and experience stays important” to forestall systemic vulnerabilities and operational failures.

Printed on Could 9, 2026