The fraud scenario that keeps Brendan Blaine Hall’s clients awake at night no longer requires a skilled hacker or a stolen password. It requires a video call. A chief financial officer takes a meeting with what appears to be the chief executive – same face, same voice, same urgency – and a wire transfer leaves the building within the hour. The call never happened. Hall, Senior Vice President at Alliant Insurance and a two-decade veteran of cybersecurity, corporate investigations, and litigation support, frames the problem not as a technology failure but as a contract one. Companies have spent years buying cyber insurance written for a world where attacks looked different. The attacks changed. Most policies did not. And the moment a company discovers the gap is usually the worst possible moment to discover it, when the money is already gone, and the claim is already filed.
Read The Policy Against Today’s Attacks
The language sitting inside most cyber policies was drafted against a threat model that assumed a human adversary doing human-speed work. Phishing emails with telltale errors. Credential theft. Ransomware deployed by people who had to break in first. Deepfake impersonation collapses that model, because nothing is technically breached at all. An employee with full authority makes a legitimate transfer based on instructions they had every reason to believe were real. Hall’s point is that the policy terms governing this scenario were written before it was commonplace, and the definitions have not caught up. “Many policies were written for yesterday’s threats,” he says. “Look closely at how yours define social engineering, funds transfer fraud, and impersonation.”
Those three definitions are where coverage lives or dies, and they are rarely read with any precision until a claim is in dispute. Social engineering coverage is often sublimited well below the main policy. Funds transfer fraud may require evidence of unauthorized access, which a deepfake call does not produce. Impersonation clauses may contemplate a forged email rather than synthetic video. Hall’s recommendation is unusually direct for an insurance professional, and it amounts to forcing the question before an insurer has any financial incentive to answer it narrowly. “Ask your broker directly how a deepfake voice or video scam would be treated,” he says. “Clear language today saves a painful dispute later.” The ask is simple, the answer is binding, and almost nobody makes it.
Build The Controls Underwriters Reward
The second half of the problem is that insurers are no longer pricing cyber risk as a blind actuarial exercise. They are pricing operational discipline, and they are increasingly able to verify it. That changes the calculus for boards who have treated controls as a cost center and insurance as the hedge against underinvesting in them. The hedge is getting more expensive for companies that cannot demonstrate basic rigor, and cheaper for those that can. “Insurers are rewarding companies that can show discipline,” Hall says. Multifactor authentication, callback verification for payments, tested backups, and a rehearsed incident response plan are the specific items he names, and the list is notable for how ordinary it is.
Callback verification deserves particular attention, because it is the one control on that list that directly defeats the deepfake scenario. An out-of-band confirmation to a known number breaks the attack regardless of how convincing the video was, and it costs nothing but process. The word “rehearsed” is doing similar work in Hall’s framing of incident response. A plan that exists as a document and a plan that has been run under pressure are different assets entirely, and only one of them shortens recovery time. The dual return is what makes the investment case straightforward. “Strong controls lead to better terms and a faster recovery when something goes wrong,” Hall says. Companies get a better premium before the incident and a shorter outage during it, which is a rare combination in risk management.
Price The Incident At Full Cost
The most expensive misconception Hall identifies is treating a cyber event as an IT problem with an IT-sized bill. The forensic investigation is only the opening expense. Legal counsel, notification obligations, business interruption, and reputational damage all follow, and several of those costs run for months after the technical remediation is complete. “A breach touches far more than your IT team,” he says. Boards that have modeled only the obvious line items will find their coverage limits adequate for the first invoice and nowhere near sufficient for the rest. The gap does not show up in a renewal conversation. It shows up in a quarterly earnings call.
Hall extends this to transactions, where the exposure is most often overlooked entirely. Acquirers conduct financial, legal, and commercial diligence as a matter of course, then inherit a target’s unexamined security posture along with its customer list. “If you’re acquiring a company, you’re also acquiring its cyber history,” he says. That history includes dormant compromises, unreported incidents, and insurance that may not transfer cleanly. His prescription is to make cyber diligence standard rather than exceptional in the deal process, which is a modest procedural change with significant downside protection. The broader argument holds across all three areas. “AI is changing how attackers operate, and your insurance strategy should evolve with it,” Hall says. Review your language, invest in your controls, and plan for the full picture. The attackers have already upgraded. The policies and the diligence checklists have not, and the interval between those two facts is where the losses will land.
Follow Brendan Blaine Hall on LinkedIn for more insights on cyber risk, AI-driven fraud, and insurance strategy for boards and executives.









