AI is a Bubble

For weeks I’ve been saying internally that this might be the summer of decision. Not only because of the ongoing economic uncertainty, but also because of the question: how will AI adoption move forward? With the release of GPT-5, i feel confirmed in my view: AI is a bubble.

Does that mean technological progress has disappointed us?

On the contrary, our expectations have even been exceeded.
But never before has the gap been so clear between technological progress, market expectations, and the actual maturity of working with generative models.

• AI-“consultants” with thousands of followers still rack up likes for prompt tips from 2023.
• Critics re-emerge claiming LLMs have no business value, even though we remain firmly on the predicted scaling line.
• And IT providers are shocked when we can prototype their latest business case in minutes with just a few prompts.

The situation reminds us strongly of the dot-com bubble. Back then, markets were overheated, buzzwords were louder than substance, and many hopes were disappointed. Yet the players who built infrastructure became today’s winners: Amazon, Google, Cloud, e-commerce.

My lesson: A bubble does not mean a technology is irrelevant. It means expectations outpace infrastructure and that’s when substance gets separated from noise.
‍

Where we stand with AI today

• Technological
  ‍GPT-5 is a clear step forward. Google’s Genie even has the potential to reshape the entire gaming sector.
• Economic
  ‍Broad adoption still lags behind. The main beneficiaries so far are solo entrepreneurs and small teams who started adapting early.
• Societal
  ‍Opinions swing between salvation promises and total rejection.

Why infrastructure is decisive

After the dot-com crash, it wasn’t the loudest startups that survived, it was those that built sustainable infrastructure and value.

For AI, the same is true today:

• Data Engineering: clean, integrated data instead of silos.
• Data Governance: security, compliance, GDPR.‍
• Integration: embedding models into real workflows, not just shiny demos

The winners after the hype will be those who establish thefoundations for AI readiness.

What companies should do now

Don’t chase every shiny new tool.

But don’t sit back and wait it out either.

Instead: invest in data products and analytics pipelines. Do AI doesn’t remain hype, but delivers real value.

How inics supports

In many conversations we notice that the connection between business intelligence, data analytics, and AI readiness is still underestimated. But in reality, it is the foundation of everything that comes next.

That’s why we help our clients with:

• Robust data products and analytics pipelines
• Governance and compliance, so AI can be deployed legally, securely, and future-proof
• Integration support, turning prototypes into real business value
• Concrete implementations – for example, integrating a data catalogue in Databricks to make data consistently discoverable, traceable, and usable for AI models

We don’t help our clients ride the hype – we help them build the foundation that lasts after the hype is gone.

„The partnership with inics transformed our BI landscape, significantly reducing costs and establishing an innovative platform for future digital and AI initiatives. We see inics as strategic digital evangelists.“

Luka Bebensee, Operations Partner - Head of Quantoo/TMG
‍

Conclusion

Yes, AI is a bubble.
But so was the internet. And just like with the internet, the winners will be those who invest in infrastructure, data readiness, and governance today.

The bubble may soon burst. The real question is: are you building the foundation for what comes after?