There were 333 ransomware attacks detected by Trellix specifically targeting critical infrastructure sectors from April 1 to September 30, 2025.

Google Gemini 2.5 Pro achieved a 59% pass rate on security tests.

97% of organizations indicated that identity-related controls influence their cyber insurance premiums or coverage terms in some way.

36% of marketplace shoppers encounter scams weekly.

In 2025, digital forgeries made up 35% of document fraud, up from an average of 29% between 2022 and 2024.

Fraud attempts peaked between 2:00 am and 4:00 am UTC.

60% of all cyber training focuses on vulnerabilities that are more than two years old.

The average containment time for simulated cyber attacks is 29 hours.

46% of organizations that filed claims reported that the incident triggering their claim was either identity-related or caused by a privileged account compromise.

86% of organizations reported that their insurers offered premium reductions or credits for their use of AI in security controls.

Anthropic’s Claude Sonnet 4.5 achieved a 50% pass rate on security tests.

77% of Chief Information Security Officers identified AI-generated phishing as a serious and emerging threat in 2025

Only 33% of cyber insurance policies cover lost revenue, and 45% cover ransomware negotiations or payment.

By 2030, more than 40% of enterprises are predicted to experience security or compliance incidents linked to unauthorized shadow AI.

53% of Defense Industrial Base members are currently using a cloud service provider to minimize their Cybersecurity Maturity Model Certification scope.

In Q3 2025, Beazley Security published 11 critical zero-day advisories, marking a 38% increase from Q2.

64% of organizations that experienced a decrease in overall cyber insurance costs in the past year credited AI adoption as a key factor.

58% of consumers have encountered ad-related malware.

60% of Defense Industrial Base members reported an increase in training staff on cybersecurity since last year, up from 37% .

In Q3 2025, the number of newly published vulnerabilities reached over 11,700, with nearly 1,800 classified as high-risk.