Cyberattack tactics are outpacing what renewal forms ask

A documented shift in how ransomware gets in puts the coverage clients think they have at risk
Cyberattack tactics are outpacing what renewal forms ask
CYBER
By Roxanne San Jose
Nov 20, 2026

Cyber insurance renewal conversations have rested on a consistent assumption for several years: show that multi-factor authentication is in place, and your insurer will look at you differently. That assumption is now a liability.

A quarterly threat report released in November 2026 by Beazley Security documented four distinct techniques attackers used in the third quarter to bypass multi-factor authentication without breaking it. These included stealing active sessions after a user had already authenticated, using device-code phishing to intercept tokens during the login process, harvesting session tokens through infostealer malware, and – in the most persistent cases – registering attacker-controlled devices to retain access after credentials were reset.

“Attackers have learned to wait until the user signs in, then steal the session,” said Josh Carolan, director of security research at Beazley Security.

Brokers advising clients on cyber insurance renewal preparation in the US need to understand what this means for their clients’ attestations before the next renewal cycle, not after a claim is filed.

Read next: Confronting the next wave of cyber threats: How brokers can bridge critical gaps

The gap between attestation and actual coverage

Most cyber insurance applications still ask whether MFA is in place. They do not ask whether it is phishing-resistant, whether it covers privileged accounts as well as standard user accounts, or whether it extends to cloud management consoles and remote access tools. As the Beazley report makes clear, the methods attackers are now using do not require them to defeat MFA at all. Underwriters need to start asking whether it is phishing-resistant, and on which accounts.

The gap between what a questionnaire records and what an attack actually encounters is a direct coverage risk. A client who attests that MFA is in place but whose implementation does not defend against session hijacking or device-code phishing may find that attestation scrutinized during the claims process.

This is a live risk in the US. The FBI’s 2025 Internet Crime Report recorded more than 3,600 ransomware complaints and US$32.3 million in reported losses, with total cybercrime losses reaching almost US$20.9 billion – a 26% increase on 2024. The agency identified 63 new ransomware variants, with Akira, Qilin, and INC ranking as the most active. The FBI has repeatedly noted that ransomware losses are significantly underreported. The pipeline from compromised session to ransomware deployment is shortening, not lengthening.

Exploitation doubles as patch volumes overwhelm teams

MFA bypass is only part of the Q3 picture. Beazley reported that the share of ransomware cases originating from an exploited vulnerability doubled from 14% in Q2 to 28% in Q3. Attackers moved away from front-door credential attacks toward internet-facing, self-hosted applications that receive less patch management attention than operating systems and endpoint software.

The volume of publicly disclosed vulnerabilities reached approximately 37,000 in Q3, up nearly 80% over Q2. At that rate, security teams face a triage problem: not every vulnerability can be patched immediately, and attackers are increasingly chaining multiple lower-severity vulnerabilities together, escalating access, and deploying ransomware through the victim’s own cloud infrastructure.

“At today’s volume, security teams simply cannot treat every vulnerability as equally urgent,” said Alton Kizziah, CEO of Beazley Security.

For brokers preparing clients for renewal, patch management documentation has become as important as MFA attestation. Clients should be able to show a risk-based patching cadence, evidence of prioritization, and a clear process for addressing critical vulnerabilities in internet-facing systems. The Beazley report noted that ransomware frequency by sector in Q3 placed healthcare at 21% of investigated cases, business services at 18%, and manufacturing and distribution at 17%.

AI agents run a breach without human direction

The Q3 report contains the first publicly documented case of a breach conducted almost entirely by coordinated AI agents. The intrusion involved approximately 270 scripts and around a dozen exploits, with the AI agents autonomously scanning for vulnerabilities, selecting attack paths, escalating privileges, and exfiltrating data.

“These agents use the same old tricks,” said Francisco Donoso, chief product and technology officer at Beazley Security. “They just find the unpatched server or the flat network faster than any human team.”

The CISA, NSA, and FBI issued a joint advisory in 2026 warning that AI-enabled cyber threats against critical infrastructure are no longer theoretical but active, with frontier AI models performing cyber-related tasks faster, at greater scale, and at lower cost than skilled human practitioners. Separately, the NAIC’s Model Bulletin on Use of Artificial Intelligence Systems by Insurers, now adopted in more than 20 states, requires insurers to maintain a written AI Systems Program with board-level accountability and risk controls – including oversight of third-party AI tools. The NAIC’s Cybersecurity Working Group has also drafted 2026 amendments to the Insurance Data Security Model Law (MDL-668) that would impose more specific third-party oversight requirements.

Cyber insurance premiums in the US have been softening even as that risk picture has darkened. That is a disconnect brokers are well-placed to explain before the market corrects.

Read next: US cyber losses hit $20.9bn in 2025: Public entities face significant cyber risk gap

The ClickFix timeline is worth raising in client conversations directly. Beazley found that ransomware typically followed a ClickFix infection by two to three weeks – long enough that the initial compromise may not be detected before the ransomware payload arrives. Clients whose incident response plans have not been tested against that kind of delayed deployment may be carrying an uninsured gap they do not know exists.

The full Beazley Security Quarterly Threat Report for Q3 2026 is available at beazley.security/insights/quarterly-threat-report-third-quarter-2026.

Free newsletter

We'll keep you up-to-date with the latest breaking news, cutting edge opinion, and expert analysis affecting both your business and the industry as whole.

Free newsletter

Our daily newsletter is FREE and keeps you up - to - date with the world of Insurance. Please complete the form below and click on subscribe for daily newsletters from IB US.