Skip to main content

Why Enterprises Are So Unhappy with Their IT Infrastructure

Kevin Cochrane
Vultr

This year, many of the cloud infrastructure contracts signed in the early days of the AI boom will come up for renewal. As the year goes on, I anticipate we'll see a significant amount of cloud vendor swapouts and multi-cloud adoption, and the reason isn't just GPU depreciation. It's because they're tired of their current cloud providers.

My company recently commissioned a survey of enterprise leaders to better understand what they need from their IT investments. The Rise Of Alternative Cloud Providers, a 2026 commissioned study conducted by Forrester Consulting on our behalf, found that nearly half of all respondents showed levels of satisfaction ranging between "not at all satisfied" and "somewhat satisfied" with their current cloud providers, for both conventional and AI/ML workloads. Legacy single-cloud models cannot sufficiently support the demands of everyday digital operations, let alone the latest advancements in AI.

For today's enterprises, modernized infrastructure is an existential priority. With the right foundation, businesses can open new revenue streams through AI tools and services, while simultaneously optimizing their existing processes to be more efficient. But with wrong-sized infrastructure, they're losing money, losing opportunity and losing ground to the competition.

What's Causing Dissatisfaction

As organizations expand their cloud services usage to accommodate heavier AI workloads, they're uncovering infrastructural weaknesses that become increasingly harder to manage at scale.

The aforementioned survey identified security and risk as the top challenge enterprises face with their current cloud providers. This high ranking is partly symptomatic of an increasingly unpredictable threat environment: the same AI innovations that have helped businesses to transform their operations are also accelerating cyberattacks.

But leaders are also skeptical of their provider's native defenses. Legacy IT infrastructure was built for a different time; enterprise leaders are understandably worried that it could be too inflexible to keep up with the pace of evolving threats.

The digital sovereignty push has also ramped up this year; data residency and encryption requirements are becoming increasingly strict, pushing businesses to seek out local cloud alternatives to remain compliant.

Behind security, the next big infrastructure challenge enterprises cited was, unsurprisingly, high costs. Between the memory shortage and stalled data center buildouts, an AI services crisis is looming. Hyperscalers are racing to expand their AI compute availability, but customers are already (unfairly) footing the bill.

As if that weren't enough, hyperscalers are already notorious for spiking costs with feature bloat and forced overprovisioning. Hyperscalers are designed for the biggest companies and governments, but not every enterprise requires that heft of hardware. Organizations end up paying for extra features simply because they come with the contract. Think of it like adding items to your shopping cart to get a discount on shipping — at the end of the day, you're still paying a lot more.

Unpredictable workloads also contribute to high computing costs. When businesses race to bring a new AI model into their processes, or have teams working with different AI tools in silos, cloud costs can spiral upward before enterprises are able to get a handle on them. Strategies like protocol tagging help, but they'll only go so far if the actual infrastructure is virtually designed to encourage overuse.

Other factors signaled organizational and design limitations. 52% of respondents cited a shortage of AI talent. 48% noted a lack of visibility into cloud infrastructure, and 46% said they were unable to scale in line with demand.

All of these factors prevent organizations from reaching the transformational stage of AI maturity — thereby preventing them from solving the business inefficiencies that are holding them back from growth.

It's Not Just the Cloud

To support next-generation AI workloads, many enterprises are considering a full-scale infrastructure rebuild that also includes edge and on-prem compute. A hybrid, multi-cloud approach would allow them to control costs through more strategic workload management across their entire IT infrastructure.

In practice, this looks like:

  • Running smaller AI models on-device to reduce latency and maximize the impact of highly specialized use cases.
  • Distributing inference across multiple GPU and CPU frameworks for improved tokenomics — a must as token-based pricing continues to rattle IT budgets.
  • Auditing for unused features to inform budget reallocation, hyperscaler contract amendments, and alternative cloud adoption.
  • Promoting more efficient prompting throughout the organization to reduce costly redundant inference.

The Solutions: Alternative Clouds, Open Source Support, and Hyperscaler Refreshes

Flexibility, scalability and security are the formula for resilient infrastructure. Most hyperscalers, with their deeply integrated stacks and sprawling feature sets, cannot offer all three. Even so, as our survey found, 54% of enterprises still plan to expand their footprint with their current hyperscalers to solve their infrastructure challenges. Uncertainty and skepticism hold them back from pursuing alternatives, but so do more practical considerations, like data migration burdens and IT team training.

The ideal solution is not to simply pick an alternative cloud and hope for the best, or to sign another hyperscaler contract with gritted teeth. Wary organizations don't have to abandon their hyperscaler outright to take advantage of alternative cloud and on-device options. Multi-cloud strategies enable workload-specific deployments; compute power becomes proportional to compute intensity, bringing down costs without sacrificing performance.

Also, infrastructure is more than just hardware. Software-supported AI compute adds robust capabilities and further promotes hardware optimization. Infrastructure with open-source compatibility allows enterprises to truly customize their stacks, whether they need additional storage, networking support, or deployment assistance. With open-source software, enterprises can get pre-built, pre-tested foundations for a fraction of the cost and the backing of highly skilled developers.

Ultimately, how enterprises solve their cloud dissatisfaction comes down to solving their biggest pain points while advancing their growth goals. Plenty will stay with their current providers, but there are too many options for unhappy customers to leave migration off the table.

Kevin Cochrane is the Chief Marketing Officer of Vultr

The Latest

Rapid AI adoption and the unique ways AI workloads operate is redefining the scope and structure of what these teams must deliver. This shift is forcing organizations to rethink how they manage scale, automation, and control, according to The State of SRE and Platform Engineering 2026, a new report from Dynatrace ...

AI is usually talked about as a software tool, but it also depends heavily on the network behind it. Whether a company is using AI for chatbots, automation, monitoring, analytics, or employee support, all of that information has to move across the network in a reliable and secure way. That means AI is not just an application decision. It is also an infrastructure decision. Before organizations rush into AI, they should ask a simple question: Is our network ready to support it? ...

Enterprise AI often lacks governed access to where business processes actually execute. Without that access, AI agents may be able to reason, but they cannot operate reliably across enterprise workflows. For AI agents to effectively carry out workflows, they will require integration-layer context and controls. Organizations can implement these prerequisites by providing AI with managed access to the middleware layer ...

Enterprise networks rarely behave the same way for very long. A routing adjustment in one region may unexpectedly alter application performance in another. A cloud migration may introduce hidden dependencies that go unnoticed until an outage occurs. All the while, the network is managed by several different teams, each of whom use different tool sets — and as a result, have different views of the network ... There’s usually an engineer who remembers why traffic fails over a certain way between sites, or which transparent firewall was added where. The problem is that human memory cannot scale alongside enterprise-scale networks ...

Ask an infrastructure team how confident they are in their ability to govern AI, and most will tell you they've got it handled. A recent survey of 406 IT decision-makers and platform engineering leaders found 86% expressing exactly that confidence. Ask the same group whether they have a formal written AI governance policy, and the number drops to 30%, according to Spacelift's Infrastructure Automation Report ...

In MEAN TIME TO INSIGHT Episode 27, Shamus McGillicuddy, EMA VP of Research, Network Infrastructure and Operations, and Parker Hathcock, EMA Research Director covering IT Service/Operations (ServiceOps), discuss observability unification in modern IT operations ... 

Virtual Private Networks became a cornerstone of enterprise security at a time when corporate infrastructure looked very different from today ... For years, this model worked well. But the architecture behind VPNs assumed a centralized corporate environment—one where the network itself was the hub of activity. In a cloud — first world, that assumption no longer holds ...

Website outages get resolved just as fast in August as they do in November. I went looking for the opposite: the summer slowdown everyone assumes is there once the people who fix things are away. It isn't in the data we collected, covering 1.8 million confirmed outages across tens of thousands of websites ...

This year, many of the cloud infrastructure contracts signed in the early days of the AI boom will come up for renewal. As the year goes on, I anticipate we'll see a significant amount of cloud vendor swapouts and multi-cloud adoption, and the reason isn't just GPU depreciation. It's because they're tired of their current cloud providers ...

There's a moment the many observability teams have experienced days into bringing a new service into production: you realize that the vendor's claims of "intelligent" behavior included a large serving of hype. Their dashboards look nice until they don't, the failure modes are a black box, and no one on the team can confidently explain why the system did what it did at 2 am. Agentic AI is about to force every Ops team to relive that moment at web-scale until they start treating these systems as the dependencies they actually are ...

Why Enterprises Are So Unhappy with Their IT Infrastructure

Kevin Cochrane
Vultr

This year, many of the cloud infrastructure contracts signed in the early days of the AI boom will come up for renewal. As the year goes on, I anticipate we'll see a significant amount of cloud vendor swapouts and multi-cloud adoption, and the reason isn't just GPU depreciation. It's because they're tired of their current cloud providers.

My company recently commissioned a survey of enterprise leaders to better understand what they need from their IT investments. The Rise Of Alternative Cloud Providers, a 2026 commissioned study conducted by Forrester Consulting on our behalf, found that nearly half of all respondents showed levels of satisfaction ranging between "not at all satisfied" and "somewhat satisfied" with their current cloud providers, for both conventional and AI/ML workloads. Legacy single-cloud models cannot sufficiently support the demands of everyday digital operations, let alone the latest advancements in AI.

For today's enterprises, modernized infrastructure is an existential priority. With the right foundation, businesses can open new revenue streams through AI tools and services, while simultaneously optimizing their existing processes to be more efficient. But with wrong-sized infrastructure, they're losing money, losing opportunity and losing ground to the competition.

What's Causing Dissatisfaction

As organizations expand their cloud services usage to accommodate heavier AI workloads, they're uncovering infrastructural weaknesses that become increasingly harder to manage at scale.

The aforementioned survey identified security and risk as the top challenge enterprises face with their current cloud providers. This high ranking is partly symptomatic of an increasingly unpredictable threat environment: the same AI innovations that have helped businesses to transform their operations are also accelerating cyberattacks.

But leaders are also skeptical of their provider's native defenses. Legacy IT infrastructure was built for a different time; enterprise leaders are understandably worried that it could be too inflexible to keep up with the pace of evolving threats.

The digital sovereignty push has also ramped up this year; data residency and encryption requirements are becoming increasingly strict, pushing businesses to seek out local cloud alternatives to remain compliant.

Behind security, the next big infrastructure challenge enterprises cited was, unsurprisingly, high costs. Between the memory shortage and stalled data center buildouts, an AI services crisis is looming. Hyperscalers are racing to expand their AI compute availability, but customers are already (unfairly) footing the bill.

As if that weren't enough, hyperscalers are already notorious for spiking costs with feature bloat and forced overprovisioning. Hyperscalers are designed for the biggest companies and governments, but not every enterprise requires that heft of hardware. Organizations end up paying for extra features simply because they come with the contract. Think of it like adding items to your shopping cart to get a discount on shipping — at the end of the day, you're still paying a lot more.

Unpredictable workloads also contribute to high computing costs. When businesses race to bring a new AI model into their processes, or have teams working with different AI tools in silos, cloud costs can spiral upward before enterprises are able to get a handle on them. Strategies like protocol tagging help, but they'll only go so far if the actual infrastructure is virtually designed to encourage overuse.

Other factors signaled organizational and design limitations. 52% of respondents cited a shortage of AI talent. 48% noted a lack of visibility into cloud infrastructure, and 46% said they were unable to scale in line with demand.

All of these factors prevent organizations from reaching the transformational stage of AI maturity — thereby preventing them from solving the business inefficiencies that are holding them back from growth.

It's Not Just the Cloud

To support next-generation AI workloads, many enterprises are considering a full-scale infrastructure rebuild that also includes edge and on-prem compute. A hybrid, multi-cloud approach would allow them to control costs through more strategic workload management across their entire IT infrastructure.

In practice, this looks like:

  • Running smaller AI models on-device to reduce latency and maximize the impact of highly specialized use cases.
  • Distributing inference across multiple GPU and CPU frameworks for improved tokenomics — a must as token-based pricing continues to rattle IT budgets.
  • Auditing for unused features to inform budget reallocation, hyperscaler contract amendments, and alternative cloud adoption.
  • Promoting more efficient prompting throughout the organization to reduce costly redundant inference.

The Solutions: Alternative Clouds, Open Source Support, and Hyperscaler Refreshes

Flexibility, scalability and security are the formula for resilient infrastructure. Most hyperscalers, with their deeply integrated stacks and sprawling feature sets, cannot offer all three. Even so, as our survey found, 54% of enterprises still plan to expand their footprint with their current hyperscalers to solve their infrastructure challenges. Uncertainty and skepticism hold them back from pursuing alternatives, but so do more practical considerations, like data migration burdens and IT team training.

The ideal solution is not to simply pick an alternative cloud and hope for the best, or to sign another hyperscaler contract with gritted teeth. Wary organizations don't have to abandon their hyperscaler outright to take advantage of alternative cloud and on-device options. Multi-cloud strategies enable workload-specific deployments; compute power becomes proportional to compute intensity, bringing down costs without sacrificing performance.

Also, infrastructure is more than just hardware. Software-supported AI compute adds robust capabilities and further promotes hardware optimization. Infrastructure with open-source compatibility allows enterprises to truly customize their stacks, whether they need additional storage, networking support, or deployment assistance. With open-source software, enterprises can get pre-built, pre-tested foundations for a fraction of the cost and the backing of highly skilled developers.

Ultimately, how enterprises solve their cloud dissatisfaction comes down to solving their biggest pain points while advancing their growth goals. Plenty will stay with their current providers, but there are too many options for unhappy customers to leave migration off the table.

Kevin Cochrane is the Chief Marketing Officer of Vultr

The Latest

Rapid AI adoption and the unique ways AI workloads operate is redefining the scope and structure of what these teams must deliver. This shift is forcing organizations to rethink how they manage scale, automation, and control, according to The State of SRE and Platform Engineering 2026, a new report from Dynatrace ...

AI is usually talked about as a software tool, but it also depends heavily on the network behind it. Whether a company is using AI for chatbots, automation, monitoring, analytics, or employee support, all of that information has to move across the network in a reliable and secure way. That means AI is not just an application decision. It is also an infrastructure decision. Before organizations rush into AI, they should ask a simple question: Is our network ready to support it? ...

Enterprise AI often lacks governed access to where business processes actually execute. Without that access, AI agents may be able to reason, but they cannot operate reliably across enterprise workflows. For AI agents to effectively carry out workflows, they will require integration-layer context and controls. Organizations can implement these prerequisites by providing AI with managed access to the middleware layer ...

Enterprise networks rarely behave the same way for very long. A routing adjustment in one region may unexpectedly alter application performance in another. A cloud migration may introduce hidden dependencies that go unnoticed until an outage occurs. All the while, the network is managed by several different teams, each of whom use different tool sets — and as a result, have different views of the network ... There’s usually an engineer who remembers why traffic fails over a certain way between sites, or which transparent firewall was added where. The problem is that human memory cannot scale alongside enterprise-scale networks ...

Ask an infrastructure team how confident they are in their ability to govern AI, and most will tell you they've got it handled. A recent survey of 406 IT decision-makers and platform engineering leaders found 86% expressing exactly that confidence. Ask the same group whether they have a formal written AI governance policy, and the number drops to 30%, according to Spacelift's Infrastructure Automation Report ...

In MEAN TIME TO INSIGHT Episode 27, Shamus McGillicuddy, EMA VP of Research, Network Infrastructure and Operations, and Parker Hathcock, EMA Research Director covering IT Service/Operations (ServiceOps), discuss observability unification in modern IT operations ... 

Virtual Private Networks became a cornerstone of enterprise security at a time when corporate infrastructure looked very different from today ... For years, this model worked well. But the architecture behind VPNs assumed a centralized corporate environment—one where the network itself was the hub of activity. In a cloud — first world, that assumption no longer holds ...

Website outages get resolved just as fast in August as they do in November. I went looking for the opposite: the summer slowdown everyone assumes is there once the people who fix things are away. It isn't in the data we collected, covering 1.8 million confirmed outages across tens of thousands of websites ...

This year, many of the cloud infrastructure contracts signed in the early days of the AI boom will come up for renewal. As the year goes on, I anticipate we'll see a significant amount of cloud vendor swapouts and multi-cloud adoption, and the reason isn't just GPU depreciation. It's because they're tired of their current cloud providers ...

There's a moment the many observability teams have experienced days into bringing a new service into production: you realize that the vendor's claims of "intelligent" behavior included a large serving of hype. Their dashboards look nice until they don't, the failure modes are a black box, and no one on the team can confidently explain why the system did what it did at 2 am. Agentic AI is about to force every Ops team to relive that moment at web-scale until they start treating these systems as the dependencies they actually are ...