95% of Companies that Invested in AI Didn't Increase Revenue - Why This Doesn't Mean the End of the Revolution, But the Beginning of the Most Important Phase

If you've been following the news about artificial intelligence over the last 24 hours, you've probably come across a series of headlines that may seem contradictory at first glance. While Google launches its revolutionary AI Mode in Brazil, a MIT study reveals that 95% of companies have not made a return of the billionaire investments in generative AI.

Paradoxical? Actually, no. We're experiencing exactly what economic theory predicts for technological revolutions - and that's excellent news.

The “Stumble” Was Expected (and Necessary)

The analysis published by S.Paulo Newspaper brings a fundamental historical perspective: Carlota Perez, author of “Technological Revolutions and Financial Capital”, demonstrates that all major technological revolutions follow a predictable pattern.

First comes installation phase - characterized by massive investments, financial manias and, inevitably, bubbles. That's exactly where we are now, with companies like Google, Amazon, Microsoft and Meta spending a staggering US$ 750 billion on data centers for AI in 2025 and 2026 alone.

Then comes the “crash” - or several stumbles - followed by the real golden age, when technology finally delivers its real transformative potential.

As Perez points out: “I've never seen a golden age happen without a crash”.

MIT Data Tells a More Complex Story

The MIT study that pointed to 95% of “failure” in AI investments deserves a closer look. When we dig deep into the data, we discover something interesting: 80% of organizations have adopted basic tools, But these tools only improve individual employee performance.

The problem isn't AI itself - it's how companies are applying the technology.

“Companies still use generative AI in a disorganized and individual way. They optimize their tasks, but this doesn't translate into a financial impact because entire processes haven't been rethought,” Vitor Elman, a member of IAB Brasil's AI committee, explained to IstoÉ Dinheiro.

In my experience supporting thousands of startups over the years, I've seen this pattern repeated with other technologies. The first wave always focuses on automate what already exists. The second wave - the one that generates real value - reimagines entire processes.

The Companies That Get It Right

Only 2 of the 9 sectors analyzed by MIT showed significant structural changes: technology and telecommunications. Not coincidentally, these are sectors that:

• Developed flexible systems that integrate into the workflow
• They created learning processes with continuous user feedback
• Rethinking entire workflows instead of just optimizing isolated tasks

The “AI Psychosis” Alert Reveals Another Challenge

Mustafa Suleyman, head of AI at Microsoft, brought up an additional concern: the phenomenon of “AI psychosis” - people developing excessive dependence on chatbots and believing in imaginary realities created by these tools.

This warning does not invalidate the potential of AI, but it does reinforce the need for it to be adopted conscious and responsible. As Susan Shelmerdine, an AI researcher at Great Ormond Street Hospital, suggested: perhaps doctors should ask patients about AI use, just as they ask about smoking or alcohol.

The point is not to avoid technology, but to use it in a balanced way, always keeping human interaction at the center.

Google and Responsible Democratization

While we face these adoption challenges, the launch of Google's AI Mode in Brazil represents exactly the kind of evolution we expect to see in the second wave.

The tool offers a more natural and intuitive experience, integrating text, voice and image in a fluid way. Most importantly, it was developed with a focus on safety and protection, especially for children - which shows maturity in the technological approach.

Bruno Pôssas, Google's global VP of search, highlighted something fundamental: several innovations are developed in Belo Horizonte, reinforcing that Brazil is not just a consumer, but a creator of technology in this revolution.

Why This Moment is a Unique Opportunity

If you lead a company or startup, this moment of “stumbling” represents the greatest opportunity of recent years. While 95% of companies have yet to figure out how to generate value with AI, you could be among the 5% who get it right from the start.

The difference lies in three fundamental approaches:

1. Systemic, Not Piecemeal Thinking

Instead of implementing AI to speed up existing tasks, ask yourself: “How can I redesign this entire process considering the possibilities of AI?”

2. Focus on Continuous Learning

Companies that thrive with AI create systems that learn and adapt with constant feedback, rather than static solutions.

3. Human Integration, Not Substitution

The most effective AI amplifies human capabilities rather than trying to replace them completely.

What to expect in the coming months

The same Folha report points out that one or more falls may occur before the golden age of AI. This means volatility in the market, re-evaluation of investments and, inevitably, consolidation of companies.

For visionary leaders, this scenario creates opportunities on three fronts:

• Talents: Skilled AI professionals may become available as companies adjust expectations
• Technology: Tools and platforms may become more accessible as competition intensifies
• Market: Companies that survive the crash emerge in a much stronger position

The Role of Education and Inclusion

One particularly inspiring piece of news in the last 24 hours: the Instituto Projeto de Vida in Uberlândia is offering free AI courses for young people.

This initiative exemplifies something fundamental: the AI revolution will only be truly democratic if it includes all segments of society. As an article on how older people are coping with AI, we need specific educational programs for different audiences.

The speed of technological change can lead to exclusion if there is no conscious effort to make AI accessible and understandable to everyone.

Regulation and Ethics: The Necessary Balance

The union dispute that Politico confronts use of AI in reporting raises crucial questions about ethical standards and responsibility.

Journalism, like other professions, needs to strike a balance between innovation and quality. AI can speed up processes and expand reach, but it can't compromise credibility - which is any outlet's most valuable asset.

Similarly, the discussion on AI in the courts shows how different sectors face the same dilemma: how to harness the efficiency of technology without compromising fundamental principles?

The future is not deterministic

The interview with Yuval Harari to Fantástico, in which he states that “AI will win and humanity will lose”, deserves careful reflection.

Harari is not being fatalistic - he is warning that we need conscious choices. As he himself pointed out, “humanity needs to reflect and act in the face of the uncontrollable advance of AI”.

The future is not predetermined. Every company that decides how to implement AI, every regulator who designs public policy, every educator who prepares the next generation - everyone is actively shaping that future.

How to Navigate This Transition

If you've made it this far, you're probably asking yourself: “How can I position my company or career to thrive in this transitional phase?”

Based on my experience supporting thousands of startups and leading innovation initiatives, I suggest five approaches:

1. invest in structured experimentation

Instead of big bets, create a portfolio of small experiments with AI. Test, learn, iterate.

2. Develop “IA Literacy” in the Team

Everyone in the organization needs to understand the basics of AI - not to program, but to identify opportunities and limitations.

3. Focus on Real Problems

Always start with the question: “What specific problem am I trying to solve?” - not “How can I use AI?”

4. Build Strategic Partnerships

No company needs to develop everything in-house. Identify partners who specialize in AI for your area.

5. Keep the Human at the Center

The best AI amplifies human capabilities. Design always with people's experience in mind - employees and customers.

The Brazilian Opportunity

Brazil is uniquely positioned to thrive in this transition. We have exceptional talent, a robust domestic market and, unlike other countries, we are still building our digital infrastructure - which allows us to incorporate AI from the outset.

Initiatives such as Google's AI Mode being developed in Belo Horizonte, free AI courses multiplying across the country, and growing responsible regulation show the maturity of the ecosystem.

We are not just spectators of this revolution - we are protagonists.

Conclusion: The Moment of Maturity

The AI “stumbles” we are seeing - from low corporate returns to warnings about overuse - are not signs of failure. They are symptoms of a rapidly maturing technology.

As with all previous technological revolutions, the first wave always disappoints in the short term and surprises in the long term. We are transitioning from the phase of manic experimentation to that of mature and responsible application.

For leaders, entrepreneurs and professionals, now is the time to position yourself strategically. Not to avoid AI, but to master it in a conscious and value-driven way.

The golden age of AI is yet to come. And those who prepare now - during the “stumbling blocks” - will be in the ideal position to prosper when it arrives.

In my mentoring and consulting work, I have helped companies and leaders navigate exactly this transition. If you want to position your organization among the 5% that are thriving with AI - rather than the 95% that are still experimenting without direction - this is the ideal time for us to talk about strategies specific to your context.