2026 AI Pulse Report

Key takeaways

• AI has rapidly transitioned from experimentation to widespread adoption, significantly influencing IT strategies and becoming the top technology priority for organizations in 2026
• Adoption velocity is exceptionally high, with 99% of organizations using or experimenting with generative AI (GenAI) and a majority actively integrating AI and machine learning services into their technology stacks
• While AI spending is surging, a substantial portion is wasted—36% of organizations report overspending on AI applications, and 14% cite wasted AI spend, highlighting the need for structured governance and outcome measurement

AI is a top IT priority for 2026

AI has moved beyond experimentation and hype cycles and into a period of historic acceleration. Across Flexera’s global research portfolio, the evidence is consistent: AI isn’t coming. It’s here, it’s multiplying, and it’s shaping IT strategy more forcefully than any technology shift in a generation.

But AI won’t prove its value without structure. Leaders need to focus on tying spend to outcomes using unit economics, setting scale-up gates tied to KPIs, measuring AI workloads like FinOps measures cloud and avoiding overbuying tools before value is evident.

The lesson from the cloud still applies: Just as hybrid cloud needed visibility and governance to deliver business value, AI requires the same disciplined approach.

AI adoption velocity is blowing past previous predictions

Adoption is accelerating even faster than the spend curves suggest. Across Flexera’s cloud and IT priorities data sets, 81% of organizations are using or experimenting with AI and machine learning (ML); 61% say they’re actively using Chatbots like ChatGPT, Claude and Gemini; and according to The Flexera 2026 IT Priorities Report, nearly all (94%) of IT decision-makers look for new ways to integrate AI into their technology stack.

But the impact isn’t limited to model usage. Heavyweight analytics that are now gaining executive attention, like Databricks and Snowflake, are now central to many enterprise AI roadmaps—and they add significant cost pressure as teams scale data pipelines, model training and inference workloads. Their growing footprint underscores how quickly AI initiatives reshape both architecture and spend.

If you asked me two years ago about my Databricks spend, I would’ve said it was just one of dozens or hundreds of software suppliers we have. Now, I can tell you that our number two supplier is Databricks and I know exactly what I spend there because we’ve spent hours working with our product management and engineering teams trying to predict what our consumption is going to look like for the next three years.”

—Jim Ryan, Flexera CEO

GenAI adoption trends: What the data says

The Flexera 2026 State of the Cloud reinforces the same acceleration curve, showing significant increases in GenAI usage. It has now moved from curiosity to critical capability, surfacing in development workflows, security and operations automation, customer-facing experiences and embedded within SaaS platforms.

Key takeaways

• AI spending is rising rapidly, along with waste. Budgets are growing, but much of the spending is hard to justify or track due to unpredictable usage and uncontrolled tool adoption
• Managing AI costs is highly unpredictable. Standard cloud cost controls don’t fit AI’s volatile workloads and shifting pricing, making financial management difficult
• Visibility is crucial. Without clear oversight of AI tools and spending, organizations risk uncontrolled costs and policy gaps. Unified visibility enables better governance and value

AI spend is surging, but so is waste

AI adoption is rising at unprecedented speed—and spend is climbing just as fast. In many organizations, waste is keeping pace. Eighty percent of leaders report that they have increased their AI investments. That being said, 36% say they spent too much on AI applications, and 14% report wasted AI spend.

This creates a challenging dynamic: AI is the top priority, budgets are expanding and yet a meaningful share of spend can’t be justified or even fully tracked. Much of the overspend stems from rapidly shifting licensing and consumption models, unpredictable usage patterns, uneven governance and proliferation of AI-powered SaaS tools acquired outside of IT’s view. In short, AI is accelerating faster than cost controls are.

AI cost unpredictability becomes one of the top cloud challenges

One aspect of the accelerating costs is that AI is changing cloud spend in ways that are increasingly hard to predict. Thirty percent of respondents listed cost management and unpredictability of AI cloud workloads as their number one challenge, 38% said it was their second top challenge and 32% ranked it as their third top challenge.

These ranks signal a structural issue: AI workloads don’t behave like traditional cloud workloads. Their consumption curves spike and shift in ways that break standard forecasting models. In practice, that means organizations need better cost attribution, policy-based guardrails for consumption, integrated insights and proactive procurement evaluation for AI-heavy platforms.

When unpredictability ranks as a universal top three challenge, it’s a clear warning sign. Without systemic visibility and governance, organizations are left reacting to AI costs instead of controlling them.”

—Conal Gallagher, CIO and CISO, Flexera

IT visibility gaps emerge as a defining risk to cost

Visibility is a critical control point to expenses. According to The Flexera 2026 IT Priorities Report, 85% of organizations say gaps in IT visibility pose a major risk, and although 92% say their employees are using AI technology well, nearly half (45%) say they don’t always know how or when their employees are using AI tools.

This is especially compounded when they can’t see who’s using AI, which SaaS tools include embedded AI, what models and providers employees interact with, where data flows and which contracts include AI-related usage or terms. For the cloud, visibility was the foundation of FinOps maturity, and for AI it’s the foundation for everything: governance, cost management, compliance, risk reduction and value realization.

As adoption accelerates and spend rises, organizations operating without unified visibility face uncontrolled experimentation, rising spend without outcomes, inconsistent policy enforcement, increased compliance exposure and cascading SaaS risk. Visibility is the prerequisite for AI control.

AI visibility is no longer just an operational concern—it’s a board‑level risk. As AI adoption accelerates, the absence of clear insight into usage, data flows and contracts creates exposure that compounds faster than most governance models can respond.”

—Conal Gallagher, CIO and CISO, Flexera

Why traditional cost management frameworks fall short

AI’s financial turbulence is revealing the limits of legacy cost management. Traditional FinOps practices—such as tagging, rightsizing or reserved instances—were built for cloud services with predictable consumption curves. AI breaks those assumptions.

All of this compounds into overspend. AI workloads spike unpredictably, AI SaaS pricing shifts without transparency and experimentation often runs outside FinOps oversight. The result is a financial model that’s far more volatile than cloud ever was—and a clear signal that AI requires new disciplines to keep cost, value and risk in balance.

The path forward: Financial clarity built on unified visibility

The organizations that get ahead won’t rely on reactive cost control. They’ll shift to proactive financial instrumentation—whether they’re using AI tools across the business or building AI capabilities for employees or customers.

Whether you’re consuming AI or creating it, the path forward is the same: visibility first, instrumentation second, governance tied to financial behavior throughout. You can’t optimize or control what you can’t see—and AI raises the stakes on both.

Key takeaways

• Rapid AI adoption is outpacing organizations’ abilities to monitor, govern and secure usage, resulting in oversight gaps and systemic risks
• Shadow AI can easily become widespread, making it increasingly difficult for organizations to maintain visibility and control, which heightens security and compliance challenges
• Effective governance, grounded in unified visibility, embedded policies and cross-functional collaboration, is essential to mitigate risks and ensure that AI delivers value without exposing the organization to liability

Shadow AI is no longer a fringe problem

Organizations are adopting AI faster than they can secure it, monitor it or even detect it—creating a widening gap between AI enthusiasm and AI oversight. The risks accumulating in that gap are no longer isolated; they’re becoming systemic. Organizations need the controls, visibility mechanisms and guardrails to protect themselves from security breaches, compliance failures, data exposure, intellectual property drift and uncontrolled AI usage.

As earlier mentioned, one sobering signal of how deeply shadow AI can penetrate the enterprise is from the 2026 Flexera IT Priorities Report, which unveiled that 45% of IT decision-makers said they don’t always know when employees are using AI.

Shadow AI emerges when employees test public GenAI tools, when SaaS platforms enable AI features by default, when browser‑based helpers slip into workflows or when unapproved plugins and assistants feel harmless. And while shadow IT was once limited to unsanctioned SaaS subscriptions, shadow AI is far harder to detect because prompts look like normal web traffic, AI features are embedded invisibly across SaaS products, vendors turn on capabilities without advance notice and many employees believe “just testing” carries no risk.

You can't defend against what you can’t see—and shadow AI is expanding faster than visibility controls can keep up. The result is a growing exposure surface with limited monitoring, limited governance and no guarantees that data, models or usage stay within approved boundaries.

Cloud-based AI security and compliance risks rise to the top

Fifty-three percent of respondents in Flexera’s 2026 State of the Cloud survey say security and compliance risks associated with cloud-based AI are their top challenge. Another 28% rank it as their second, and 20% rank it as their third top priority. This makes security and compliance the single most consistently cited AI challenge—outpacing cost, talent and operational hurdles.

AI systems introduce entirely new exposure points. They process unprecedented volumes of sensitive data, generate content that may contain regulated or proprietary information, create data flows that bypass established controls and operate with opaque model behavior that’s difficult to assess. Security teams are now expected to safeguard:

This isn’t just a technical challenge; it’s a governance gap. And as AI adoption accelerates, that gap is widening faster than traditional security models can respond.

Cloud‑based AI introduces new security and compliance risks because it operates beyond many existing guardrails. Without unified visibility and governance, those risks scale faster than most organizations realize.”

—Conal Gallagher, CIO and CISO, Flexera

External research reinforces the urgency

A 2025 Gartner survey found that 69% of cybersecurity leaders suspect or have evidence of employees using unauthorized public GenAI tools, leading to risks like IP loss, data exposure and security gaps; Gartner further predicts that by 2027, 40% of enterprises will face a security or compliance incident tied to shadow AI. Additional Gartner research shows AI-related information governance risks and shadow AI are rapidly climbing senior risk leaders’ priorities, echoing Flexera’s findings that AI adoption is outpacing organizations’ ability to govern it.

Governance is the hinge: Visibility, then control—then trust

The highest‑performing organizations aren’t avoiding AI risk; they’re controlling it by treating governance as a core operating system, not an afterthought. Their frameworks are lighter, more operational and built for real‑time oversight. In order to create governance with authority—not governance in theory—organizations need effective visibility, policy embedded in the workflow, secure data governance, AI and compliance alignment and cross-functional governance. They also need AI model security—yet another aspect of governance increasing in importance. That includes how to keep models safe from prompt injection, model poisoning, adversarial attacks and more.

The cost of getting governance wrong

Without governance and visibility, organizations face escalating exposure: data leakage, regulatory penalties, intellectual property loss, untracked spend surges, model misuse, inconsistent outputs and uncontrolled contractual liabilities. Each of these issues is damaging on its own, but together they form a material organizational risk that compounds as AI adoption accelerates.

Key takeaways

• To realize lasting AI value, organizations must scale responsibly by establishing strong governance, continuous measurement and clear alignment with business outcomes
• Measurement, visibility and financial controls are foundational—without them, even widespread AI adoption fails to deliver meaningful or measurable ROI
• Successful teams prioritize clarity, embed governance early and close visibility gaps to ensure that every AI deployment is accountable, traceable and drives tangible business results

The challenge now is moving from AI’s enormous promise to measurable results. The key isn’t just investing or experimenting—it’s scaling responsibly, tracking impact rigorously and establishing clear visibility from the outset. Without these foundations, even widespread AI deployment can fall short of driving meaningful business outcomes.

The core problem: AI adoption is high, but value realization is low

Across industries, leaders are rapidly integrating AI, but far fewer are actually capturing value. The issue isn’t a lack of usage—it’s a lack of guardrails around measurement, visibility and governance.

Cloud-based AI introduces new risks because it operates beyond many existing guardrails, and organizations don’t have the measurement in place they need. Without unified measurement, visibility and governance, those risks scale faster than most organizations realize.”

—Conal Gallagher, CIO and CISO, Flexera

As you may have realized from previous sections, data from Flexera’s research mirrors this pattern: AI investment is rising, experimentation is widespread, governance and visibility gaps persist and shadow AI continues to accelerate. Organizations are integrating AI tools, but they still struggle to connect usage to business KPIs, isolate which deployments drive ROI or measure whether initiatives deliver meaningful outcomes. This isn’t a technology failure. It’s a measurement failure.

Scaling AI responsibly: The bridge between ambition and outcomes

Responsible AI scaling means organizations don’t jump from pilot to full rollout without first establishing visibility, governance, cost discipline, measurement frameworks and operational guardrails. These foundations ensure AI is deployed not just ethically and securely, but also effectively.

AI only creates value when it can be seen, controlled and measured. FinOps for AI connects financial insight, operational visibility and governance into a single system of record.

Key takeaways

• Each industry faces unique AI challenges, but overcoming visibility, governance and measurement gaps is key to unlocking value
• Organizations achieve lasting AI success by treating it as a disciplined, measured and cross-functional operational practice
• Responsible innovation means scaling AI only after proven pilots, ensuring costs and risks are well managed

AI value maturity: What “good” looks like across industries

Organizations that consistently extract value from AI share a common pattern: They treat AI as an operational discipline, not a one‑off project. AI value isn’t left to chance—it’s deliberately engineered. It all starts with total visibility: those who track every AI model, SaaS tool and cloud workload ensure nothing escapes their view. This transparency lays the groundwork for thoughtful financial management and ongoing optimization.

But visibility isn’t enough. Governance shouldn't be siloed in one department. Security, IT, finance, legal and data teams collaborate, turning governance into a dynamic process that guides innovation and keeps everyone accountable. Instead of slowing things down, this approach makes sure every new AI initiative is aligned with broader goals.

The best organizations also set clear metrics tied to real outcomes from the start, versus waiting until the end to measure success. Measurement is ongoing, enabling teams to spot opportunities and address issues before they become problems. These teams scale cautiously, only moving forward when pilots deliver proven value and risks are well understood. This disciplined growth ensures that costs stay manageable and benefits are realized, reinforcing smart financial practices. Experimentation is encouraged, but always within transparent boundaries. Teams are constantly refining their approach and making optimization an everyday pursuit rather than a one-off task.

Organizations are adopting AI at extraordinary speed, but its benefits aren’t guaranteed. Many still lack the visibility, governance and measurement needed to manage it. The hype cycle may rise and fall, but the operational pressure created by AI is only increasing.

Leaders everywhere must shift from AI adoption to AI accountability. That means understanding where AI is used, how it behaves, what it costs, the risks it introduces and the outcomes it produces. The organizations that lead won’t be the ones that adopt the most tools—they’ll be the ones that instrument AI with discipline. They’ll maintain unified visibility across cloud, SaaS and embedded AI features. They’ll govern AI with the same rigor they apply to security and compliance. They’ll treat cost unpredictability as solvable. They’ll measure AI outcomes with KPIs tied to productivity, accuracy, cost avoidance and improved experience. And they’ll scale AI intentionally—only when value is proven and risk is controlled.

The insights in the Flexera 2026 AI Pulse Report are drawn from Flexera’s flagship research reports, primarily the Flexera 2026 IT Priorities Report and the Flexera 2026 State of the Cloud Report. These studies represent thousands of data points collected from IT, cloud, finance, security, procurement and IT asset management (ITAM) leaders across global enterprises.

The perspectives in this report are informed by Flexera’s global research and by operational signals observed across cloud optimization and financial management environments. Recent acquisitions of Chaos Genius and ProsperOps further expand this view, surfacing earlier and more granular signals of AI-driven volatility and cost behavior.

Reuse

We encourage the reuse of data, charts and text published in this report under the terms of this Creative Commons Attribution 4.0 International License. You’re free to share and make commercial use of this work as long as you provide attribution to the Flexera 2026 AI Pulse Report as stipulated in the terms of the license.

FAQ