Business Insurance
Cyber Risk Isn’t Just an IT Problem Anymore.
In 2023, MGM Resorts wasn’t hacked in the traditional sense. It was socially engineered. The attackers didn’t break in, they were let in. The threat group Scattered Spider researched an MGM employee on LinkedIn, then called the company’s IT help desk posing as that employee. No malware. No intrusion alert. Nothing for the security team to catch, because there was nothing to catch. Just a convincing enough impersonation that the help desk reset the employee’s authentication credentials and handed over direct access to MGM’s systems. Casinos were down for days, and the incident is estimated to have cost the organization roughly $100 million.
“What Am I Paying My IT Guy For?”
Current IT infrastructure is the most common pushback when a client questions why they need cyber insurance at all — and it’s the wrong instinct.
It’s rarely the technical systems that fail. It’s the human element, the instinct to trust, that fails. Most modern cyber events exploit human trust rather than technical vulnerabilities. Often, the very businesses a company relies on can be the root cause of a cyber event. That is the shape cyber risk takes today, and it’s a long way from the archetype of a masked hacker in a basement that still shapes how a lot of boardrooms picture the threat.
The criminal side of cyber risk has professionalized fast. Ransomware used to be a smash-and-grab. Now it runs like a business line, complete with affiliate networks that split the payout with whoever gained the initial access, and negotiation desks that treat a victim company the way a collections agency treats a debtor: calmly, methodically, and with full knowledge that the company can’t afford not to pay.
The people carrying out cyber-attacks increasingly don’t need deep technical skill. They need a foothold they can buy from someone who already has one, and a playbook. Artificial intelligence is compressing that timeline further: a small group can now run reconnaissance, phishing, and negotiation at a scale that used to require the larger, more sophisticated crews of 2020 and 2021.
The Impact Of AI On Cyber Risk.
AI is part of the reason why business email compromise and social engineering have overtaken the cyber risk climate, shifting away from the dramatic network breach. Someone on the finance team authorizes a transfer, resets a password, or types credentials into a page built to look exactly like the real login screen, and the company is out real money before a single piece of malware ever enters the picture.
This means that cyber policies underwritten years ago with old-school hacking in mind can have a real blind spot, because the loss didn’t come from a system being breached. It came from a person acting on instructions that looked completely legitimate.
AI is compressing that same timeline. Where a breach used to take a full team, one person can now run reconnaissance, phishing, and negotiation against a dozen targets at once. And the old advice to watch for bad spelling or grammar as a tell doesn’t hold anymore, AI writes cleaner than most of us do.
Systemic Cyber Risk.
Then there’s the exposure a company inherits without ever seeing it coming: a breach at a cloud host, a payroll vendor, or a software provider three steps removed from the customer relationship. A company can run a tight ship internally, pass every audit, and still take a direct financial hit because a trusted vendor did not. Risk now travels through the entire web of relationships that a business depends on, not just the perimeter it controls.
The 2024 CDK Global attack is a clean illustration. CDK is the software platform most North American car dealerships run their finance and service operations on. When it was breached, thousands of dealerships went down because they all depended on the same vendor. Many lost the most profitable part of their business, service and repair work, for weeks. A company’s biggest cyber vulnerability, in other words, can belong to someone else entirely.
Cyber As Executive Risk vs. An IT Problem.
In the midst of the evolving cyber risk landscape, underwriting itself has gotten sharper, and this is where the conversation stops being an IT problem and becomes a board-level one.
The questions insurers are really asking has shifted too: not just how likely a company is to have an event, but how quickly it can detect, respond to, and contain a cyber event once it happens. Five years ago this might have required a short questionnaire and a check. Today, insurers want specifics: is multi-factor authentication mandatory across the organization, are backups tested and kept offline, does the incident response plan exist as something the executive team has rehearsed or just a document filed away after a compliance exercise? Companies that can’t answer these questions are seeing higher premiums, tighter sub-limits on ransomware, or in some cases no quote at all. The market is now grading a company’s security posture directly, and that grade shows up on the balance sheet.
The legal and reputational exposure layered on top of a cyber incident can further raise the stakes. A breach can trigger notification obligations in every state where an affected customer lives, each on its own clock. For public companies, a material incident must be assessed and, if it clears the bar, disclosed to investors on a timeline measured in days, not the quarter or two it might once have taken to fully understand what happened. Boards that used to treat a cyber incident as an operational matter to be briefed on after the fact are now expected to have governance in place before anything happens, because regulators and investors alike are asking what the board knew and when.
That governance question is bigger than the IT team. In underwriting meetings, the person best positioned to advise on governance policy and cyber risk management is often legal counsel, not the CISO alone and companies that can’t show insurers, regulators, or their own board that legal is part of the response plan are increasingly finding that gap costs them, too.
Why A Detailed Policy Review Is Warranted.
Cyber risk management isn’t a single threat to defend against. It’s a moving target, shaped by how criminals monetize access, how deeply a company’s operations run through vendors it doesn’t control, and how convincingly it can show an insurer, a regulator, or its own board that it would catch a problem early rather than discover it in a headline.
Five years ago we used to ask, “how likely is this company to have an event?” We could answer that easily based on its size and the type and amount of data it carried. The question now is, when a cyber incident happens, how quickly can you respond, detect and mitigate?
The policy that got renewed on autopilot three or four years ago was built for a different, calmer version of cyber risk. Whether it still holds up is worth asking directly, on its own terms, rather than assuming the answer because nothing has gone wrong yet