Where I see AI market today and why I believe the moment for accelerated growth and adoption is only a few years away?

I've been getting questions for a while about where I think the AI market is and how it will evolve. The question is not just theoretical – the answer also defines the right time to launch investment projects dedicated to this area.

Of course, the answer depends on the risk profile and investment portfolio of each entity. If we are talking about experimentation, market learning and technological innovation, the time is now! – and in fact, this stage has already begun a few years ago.

However, I will continue to refer to growth investments, after validating technologies and business models.

Two fundamental landmarks

To explain where I think we are now, I refer to two classic theses:

📖 Crossing the Chasm – Geoffrey Moore – the theory that shows how difficult it is for a technology to move from early adopters (enthusiasts) to early majority (pragmatic users, who only adopt when there is a clear ROI).

📊 Gartner Hype Cycle – the model that describes the stages of perception of an innovation:

• Technology Trigger – the moment when innovation occurs

• Peak of Inflated Expectations – hype and exaggerated promises

• Trough of Disillusionment – disappointment, when reality doesn't keep up

• Slope of Enlightenment – the first solid and scalable applications

• Plateau of Productivity – maturation and widespread adoption

  
  

Where are we now

Today, in my opinion, AI is somewhere between crossing the chasm and trough of disillusionment. The initial enthusiasm (2022–2023) led to a huge wave of startups and funding, but many quickly died out.

At present:

• Investment funds have become more selective and are looking for real product-market fit.

• Big companies experiment, but they demand measurable ROI, not just “cool demos.”

• The infrastructure (cloud, GPUs, open-source models, regulation) is consolidating, but it is still far from mature.

In other words, the hype has passed, the disappointments are visible, and the market is entering a filtering phase.

What's next?

If we look at the history of other technological waves (cloud, mobile apps, cybersecurity), we see a pattern: the initial hype is followed by a few years of disappointment and consolidation, then the first solid applications appear and enter the mainstream.

In all these cases, the transition from hype to early majority generally took 3–5 years.

That's why I believe that AI will also enter the pragmatic adoption phase (early majority) between 2027 and 2029. That's when we will see companies that truly scale, on validated business models and with clear ROI.

Conclusion

This is also the reason why I chose to see the launch of a dedicated AI fund – NextAi Capital – as a project for the future, not the present. I believe that the optimal time to enter the market will only be when the hype has settled and the real colors of accelerated growth will appear.

I wrote more about the model I envisioned for this fund in this article about NextAi Capital .