Most companies have a reasonably good idea of who their vendors are. Procurement has contracts, finance knows where the money is going, security has questionnaires and assessments, and someone probably maintains a list identifying which vendors have access to sensitive information or important systems. What is often much less clear is which of those relationships the business has become dependent upon to the point that losing one could materially affect its ability to operate.

A recent Insurance Business article describes a growing concern in the London cyber insurance market about how long insurers can continue providing broad dependent business interruption coverage as more companies rely on the same technology providers. Dan Johnson, cyber class underwriter at AEGIS London, explains that large organizations can commonly obtain coverage across a broad technology supply chain today, but the underlying concentration of risk is becoming increasingly difficult for insurers to ignore.

The problem is fairly straightforward. Modern businesses rely heavily on external technology providers, and many of them rely on the same ones. A serious event affecting a major cloud platform, software provider or infrastructure service could therefore create business interruption losses across a large number of insured companies at the same time. From the perspective of an individual business, this looks like a third-party dependency. From the perspective of an insurer with thousands of policyholders, it can become a systemic event.

We have already seen enough examples to understand the problem. The CrowdStrike outage in 2024 demonstrated how quickly an issue involving a widely deployed technology could affect airlines, healthcare organizations, financial institutions and other businesses around the world. AEGIS London has used that event as one example of the broader accumulation problem facing cyber insurers, arguing that interconnected technology may represent one of the most difficult systemic risks the cyber market needs to address. Their analysis of the issue is available in Cyber's Biggest Problem.

What makes the current discussion particularly interesting is how insurers might respond. According to Insurance Business, one possibility is that dependent business interruption coverage eventually becomes more focused on a company's most important vendors rather than broadly covering the entire technology supply chain. An organization might believe it has hundreds of critical vendors, while an insurer could eventually be prepared to provide meaningful dependency coverage for only a much smaller number.

That would force a conversation I think many organizations should already be having. If you had to identify the ten vendors whose failure would create the greatest threat to your ability to continue operating, could you do it with confidence?

When Vendor Risk Becomes Business Risk

The answer is probably harder than it sounds because most third-party risk programs weren't designed to answer that question. They were developed primarily to evaluate the risk a vendor introduces into the organization. Security teams assess controls, privacy teams review how information is handled, procurement evaluates contractual obligations, and legal considers liability. Those activities are important, but they don't necessarily tell management what happens to the business when the vendor disappears.

A vendor can have strong security controls and still represent a significant operational dependency. Another vendor can have a large contract without being particularly important to the company's ability to continue operating. The amount of money being spent with a supplier doesn't necessarily tell you how much the business depends on it, and a traditional security rating doesn't answer that question either.

Aon has made a similar observation in its work on third-party cyber risk, noting that organizations can underestimate the financial consequences associated with smaller suppliers when their actual role in the business isn't well understood. This is one of the reasons I think third-party cyber risk needs to move beyond vendor assessment and become part of a broader discussion about how the business operates.

Consider a specialized software provider supporting a relatively narrow business process. The contract may be small enough that it receives limited executive attention, and the provider may perform well in a security assessment. If that software happens to support a process that cannot easily be performed another way, however, its operational importance can be far greater than its contract value suggests. The organization may not recognize that dependency until the service becomes unavailable and people begin discovering how many other activities relied on it.

Critical Means Something Different When Operations Stop

This is where the difference between vendor risk and operational dependency becomes important. Vendor risk asks whether doing business with a particular provider creates an acceptable exposure. Business Survivability also needs to understand what happens when a capability the company depends upon is no longer available.

Time becomes an important part of that discussion because dependency isn't always absolute. A business may be able to operate without a particular service for several hours with little consequence, while losing the same service for several days creates a serious problem. Another provider might support a transaction or operational process where even a short outage begins affecting customers almost immediately. Understanding the dependency therefore requires more than assigning the vendor a high, medium or low criticality rating. Management needs to understand how the consequences develop as the outage continues.

Why Insurers Are Paying Attention

Cyber insurance is beginning to expose this issue because dependent business interruption coverage is designed to address some of the financial consequences when a cyber event at a third party disrupts the insured company's operations. The coverage makes sense because an organization doesn't need to experience a direct attack to suffer a significant cyber loss. If an essential provider experiences an incident and the business can no longer perform an important function, the financial effect can resemble the consequences of an attack against the company itself.

The challenge for insurers is that these risks aren't independent. Thousands of companies may depend on the same cloud platform or technology provider, which means a single incident can produce losses across an insurer's portfolio at the same time. That concentration changes the economics of the coverage even when every individual policyholder appears to present a reasonable risk.

Aon's recent cyber and technology E&O market outlook reflects that concern. The current market remains favorable for many buyers, but insurers are paying closer attention to critical vendor dependencies and systemic events. Aon also warns that a significant concentrated loss involving a widely used provider could change market conditions quickly.

The policy itself can create another layer of complexity. Gallagher's 2026 Cyber Insurance Market Outlook notes that insurers continue to adjust language around supply-chain exposures. Coverage can depend on the relationship between the insured and the affected provider, and limitations involving particular technology providers may apply. Waiting periods and other provisions can also affect how much of an interruption ultimately becomes an insured loss.

For a business owner or executive, this is where the discussion needs to move beyond whether dependent business interruption coverage exists. The more useful question is whether the organization understands the dependencies it is asking the insurer to cover.

The Vendor You Know May Not Be the Real Dependency

That becomes even harder because the vendor with whom the company has a contract may not be the provider upon whom the business ultimately depends. A software application may run on a major cloud platform. A business process provider may rely on another software company. A payment service may have its own infrastructure dependencies. The company may believe it has diversified across several vendors while those vendors ultimately depend on the same underlying technology.

The visible vendor relationship can therefore hide the actual concentration of risk. What appears to be a diversified supply chain from a procurement perspective may contain a single point of failure several layers below the relationships the company directly manages.

AI Is Adding Dependencies We Haven’t Accounted For

AI is adding another dimension to this problem. Organizations are introducing AI capabilities into existing applications and business processes at a remarkable pace, but they won't always have a direct relationship with the AI provider underneath those capabilities. An application the business has used for years may introduce AI functionality that depends on an external model provider, additional cloud infrastructure or another service the customer has never evaluated directly.

Over time, that AI capability can become part of the way employees perform their work. What began as a convenience can become an operational dependency without anyone deliberately deciding that the business should depend upon it. AEGIS London's Johnson specifically includes AI providers in the broader concentration issue facing cyber insurers because they are becoming another important layer in an increasingly interconnected technology ecosystem.

This is where I think the conversation around AI risk needs to mature. There is a great deal of attention being paid to what an AI system might do incorrectly, how data is being used and whether employees are using approved tools. Those are legitimate concerns, but they don't address what happens when an AI capability becomes important to the operation and then becomes unavailable. As AI becomes embedded into ordinary business processes, understanding that dependency becomes part of understanding the business itself.

Insurance Can Cover the Loss, Not Restore the Business

Insurance can help address the financial consequences of these failures, but insurance doesn't eliminate the underlying dependency. A dependent business interruption policy may reimburse some portion of a covered loss, but it cannot restore the provider's systems or make the provider recover more quickly. It also doesn't give the company an alternative operating process if nobody developed one before the incident occurred.

That distinction is central to how I think about Business Survivability. The objective isn't to eliminate third-party dependency because modern businesses simply cannot operate that way. The objective is to understand where the business has become dependent enough on another organization that its failure could materially affect the company's ability to continue operating, and then make a conscious decision about how that risk should be managed.

In some cases, that may mean creating an alternative provider or operating process. In others, the organization may determine that redundancy is economically unrealistic and choose to accept part of the risk while transferring some of the financial consequences through insurance. The right answer will depend on the business and the dependency involved. What matters is that management understands the decision it is making.

Brown & Brown's current cyber market guidance makes a related point by encouraging organizations to plan for systemic disruptions and develop operational contingencies that extend beyond insurance coverage. That is an important distinction because financial recovery and operational recovery aren't the same thing. A claim payment can help absorb a loss after an event, but it doesn't necessarily keep the business operating while the event is underway.

Start With How the Business Actually Operates

If I were trying to identify an organization's most important third-party dependencies, I wouldn't begin with the vendor inventory. I would begin with the business and identify the capabilities that have to continue for the organization to remain viable during a serious disruption. From there, I would trace the technology and external services those capabilities depend upon, including dependencies several layers removed from the company's direct contractual relationships.

That exercise may produce a very different list from the organization's traditional inventory of critical vendors. It may also expose places where several supposedly independent business capabilities ultimately depend on the same provider. Those are the kinds of dependencies that matter when thinking about survivability because they reveal where one failure can affect much more of the organization than management expected.

A Question Management Should Already Be Able to Answer

The cyber insurance market may continue offering broad dependent business interruption coverage, or it may gradually move toward more restricted approaches as insurers become more concerned about concentration. We don't yet know exactly where that market will settle, and the current Insurance Business reporting describes possibilities being discussed rather than a market-wide change that has already occurred.

What matters to me is what the discussion reveals about the businesses being insured. If an insurer asked your company to identify the ten vendors whose failure would have the greatest impact on your ability to continue operating, the organization should already know the answer. Not because an underwriter may eventually ask the question, but because understanding those dependencies is part of understanding how the business actually works.

The worst time to discover that a vendor has become essential to the survival of your business is when that vendor is no longer available.