AI investment is rising sharply, but real spending patterns suggest most organisations are still experimenting rather than redesigning how they operate.

I recently came across an article in The Next Web called “The People We Left Behind: Tech Layoffs, AI Hype, and a Misplaced Future”, and it gave me a lot to think about.

The article highlights the uncertainty many people feel as companies shift toward AI-driven productivity. This conversation is especially strong in the US tech sector, but similar concerns are starting to appear in Europe and other regions, though often with a more careful approach.

Many people talk as if we are already in the middle of a major transformation, with AI changing the way we work. However, when I review the Soldo 2026 Spend Index data, the reality seems more subdued.

Soldo analyses real transaction data from tens of thousands of businesses, focusing on actual spending choices rather than opinions or predictions.

If AI had already become a structural change, we would see investment patterns showing long-term commitment and company-wide integration. The data, however, shows we have not reached that point yet.

The narrative: AI as structural disruption

The big-picture story is strong, with economic forecasts, leaders, and the media often describing AI as a major turning point instead of a slow change.

The IMF has estimated that AI could affect nearly 40% of jobs globally. McKinsey projects that generative AI alone could add trillions of dollars annually to the global economy. Public company earnings calls increasingly describe AI as a core priority rather than an innovation experiment.

At the same time, ongoing layoffs in some areas of the tech sector are often described, directly or indirectly, as the result of AI-driven efficiency gains.

This suggests that the transformation is already structural, but actual spending patterns tell a different story.

What structural transformation looks like in spend

In finance, real transformation eventually appears in the way money is spent. When a shift is structural, we typically see:

• Multi-year enterprise contracts
• Vendor consolidation
• Clearly defined budget lines
• Clear executive ownership of ROI
• Budget reallocation

We saw this happen with cloud infrastructure, ERP systems, and cybersecurity. Structural change becomes clear in spending when companies move from trying things out to making real commitments. So the question is whether today’s AI spending looks more like investing in infrastructure or just testing out new tools.

What the Soldo 2026 Spend Index shows about AI investment

AI spending is clearly rising, and the growth is significant. The Soldo Spring Index 2026 shows that average spending on the ten most popular AI tools went up by 77% year-on-year, from €1,205 to €2,128 per company in 2025. Total spending on these tools increased by 175%, compared to 2024. Still, the amount each company spends is modest compared to traditional infrastructure or core systems. This suggests that many organisations are scaling up AI gradually instead of making it a central part of their operations.

However, the details of this spending are important.

Based on the Soldo Spend Index patterns, AI spend today is predominantly:

• Subscription-based rather than infrastructure-based
• Distributed across departments rather than centrally consolidated
• Short-term and flexible rather than locked into multi-year contracts
• Fragmented across multiple vendors
• Included in operational budgets rather than long-term investment

All these factors show that current AI spending looks more like trying out new SaaS tools than overhauling core systems and processes. The trends within AI tools support this: in early 2025, spending on specialist tools like Cursor (+994%) and Anthropic (+489%) grew rapidly, while spending on general-purpose large language models grew more slowly, and some platforms even saw declines. This variety suggests that companies are experimenting with new uses rather than settling on a single enterprise solution.

In short, AI spending seems forward-looking and experimental, but it is not yet a fully established part of company budgets.

Trying out tools in daily work is different from changing the way a business operates.

A maturity curve perspective

How companies spend money often shows their stage of maturity more clearly than their public statements.

Most transformative technologies follow a recognisable curve:

• Exploration phase
  • Tool experimentation
  • Decentralised adoption
  • Curiosity-led use.
• Efficiency phase
  • Productivity gains
  • Workflow acceleration
  • Cost optimisation.
• Structural phase
  • Process redesign
  • Systems integration
  • Operating model change
• Strategic phase
  • New revenue streams
  • Sustainable competitive advantage

Looking at spending patterns, most organisations are still in the first two phases. Growth has been especially strong in highly regulated or complex sectors, such as financial services (+441%) and manufacturing (+418%). This shows that more industries are adopting AI, but it is still about speeding up adoption rather than fully integrating it.

AI is clearly helping workflows and making many processes more efficient, but in most cases, it has not yet become a core part of how companies operate at scale.

Still, public discussions often suggest we have already reached the final stage.

The timing gap: restructuring vs realised transformation

When layoffs are explained as being due to expected AI productivity gains, it assumes those gains are already predictable, measurable, and scalable. But if most AI investment is still operational and experimental, these productivity gains may not yet be fully built into company structures.

This creates a gap between what people expect and how companies are actually investing, especially when productivity gains are expected before they are fully realised.

In some cases, companies are reorganising in hopes of a transformation that is still being tested.

This is not meant to question AI’s long-term impact, but to point out what current spending says about how committed organisations are right now.

Spending patterns are often the clearest sign of how committed an organisation really is.

The CFO lens: why caution is rational

From a finance perspective, long-term funding follows when something becomes:

• Predictable
• Governable
• Measurable
• Scalable

Until companies can confidently predict ROI, have strong governance, and clear integration plans, funding will likely stay focused on day-to-day operations.

This is not about being sceptical. It is about careful financial management based on accountability and long-term responsibility. Finance does not resist change. It supports change when there is enough evidence for a long-term commitment.

What to watch for

When AI becomes truly structural, we will likely see:

• Vendor consolidation rather than fragmentation
• Longer contract durations
• Centralised governance frameworks (especially in highly regulated industries)
• Dedicated AI infrastructure budgets
• Clear executive accountability for measurable ROI

Real change eventually becomes clear in how money is spent and organised over time. For now, spending data shows that many organisations are still experimenting rather than making major changes to their operations.

Recognising this difference can help us move forward with more clarity and fewer assumptions.