Cyber warfare has entered a new phase where AI is accelerating both the speed and complexity of attacks. According to Armis Cyberwarfare Report 2026, organizations globally are increasingly exposed while a dangerous overconfidence in their own security is growing.
Armis , the cyber exposure and security management company, warns that cyber warfare threats have reached a global pressure cooker moment. As new technologies accelerate cyber operations and geopolitical tensions escalate, attackers are increasingly targeting the infrastructure, information and systems that underpin global stability.
”Geopolitical tensions, AI acceleration and unresolved security vulnerabilities are colliding and bringing cyber warfare to a boiling point,” says Nadir Israel, CTO and co-founder of Armis. ”Cyber warfare is now a constant state; attackers are moving at machine speed, while too many organizations are still trying to defend themselves with assumptions and structures built for a completely different threat landscape. Organizational leaders must heed the call and immediately improve their proactive cybersecurity operations before it is too late.”
New results from Armis Lab's fourth annual global cyber warfare report, A World Under Pressure: Cyberwarfare in an Age of AI-Fueled Escalation , shows how quickly the pressure is building. Nearly eight in ten IT decision-makers (79 %) globally are concerned that national actors are using AI to develop more sophisticated and targeted cyberattacks, up from 73 % last year, while two-thirds (67 %) believe that misuse of emerging technologies will increase the likelihood of unforeseen damage to civilian infrastructure during cyber conflicts.
Despite mounting pressure, there is a false sense of confidence among organizations that remains surprisingly high. Nearly eight in ten (79 %) IT decision makers say their organization is prepared to handle a cyberattack, while 76 % believe they are ready to mitigate AI-driven threats. This confidence is increasingly disconnected from reality.
More than half (54 %) of organizations report having already suffered an AI-generated or AI-led attack in the past 12 months, and half (50 %) admit they still haven’t been able to adequately secure their environment following the attack. Meanwhile, two-thirds (66 %) agree that organizations continue to underestimate the resources required to defend against AI-driven threats, revealing a growing gap between perceived preparedness and actual resilience.
”False confidence in the absence of contextual intelligence is a force multiplier for the adversary,” said Michael Freeman, head of threat intelligence at Armis. ”In an era of AI-accelerated threats, nation-state actors are exploiting the gap between a defender’s perceived security and their actual exposure. Effective cyber exposure management doesn’t just provide a ‘view’; it provides the foundational insight needed to eliminate blind spots and preemptively harden the attack surface before the first attack is even launched.”
Additional key global findings from this year’s report include:
- 69% of IT decision-makers agree that a global reliance on AI will intensify geopolitical efforts in cybersecurity.
- 68 % also agrees that the weaponization of AI will make cyber conflicts a more persistent feature of global geopolitics.
- 64 % says that emerging technologies are making it harder to distinguish between espionage, cybercrime and acts of war.
- 74 % believes that cyberwarfare attacks will increase and be directed at institutions that represent free press and independent thought.
- 52 % says ransomware payouts now exceed their annual cybersecurity budget.
- Average global ransomware payout in 2025: $11,610,000 / £7,710,000 / €8,440,000
Armis Cyberwarfare Report 2026 is based on a study of more than 1,900 global IT decision makers and proprietary data from Armis Labs. Read the full report from Armis, including a detailed regional and industry-specific breakdown, here .
Read the blog to delve into these global findings and the key lessons for organizations. here .
For a closer look at the findings in this report that captures the perceptions of IT decision-makers within US federal governments, read the blog here .
Swedish
Danish
Norwegian
Finnish
English
German
Dutch