Brazil May Be the First to Regulate AI Agents While 80% of Companies Still Underutilize the Technology — The Innovation Paradox

We are living in a fascinating and contradictory moment in the Brazilian artificial intelligence scene. As our country positions itself to potentially be the first in the world to regulate “AI agents” — a definition that could completely change how we think about technological responsibility — we are discovering that 80% of Brazilian companies have already adopted generative AI, but 75% still use it very little.

This paradox reveals something profound about our relationship with innovation: we are quick to embrace the new, but slow to realize its true potential. Over the past 24 hours, news about AI in Brazil has painted a picture that mixes regulatory pioneering, practical challenges, and untapped opportunities.

Brazil at the forefront of global regulatory developments

According to a report by Poder360, the bill to regulate AI in Brazil, which has already passed the Senate and is being processed in the Chamber, maintains an innovative definition of “AI agents” that encompasses developers, distributors and applicators involved in the value chain of AI systems.

This approach is revolutionary because it treats AI not as an isolated tool, but as a ecosystem of shared responsibilities. It's like looking at a car and realizing that the responsibility lies not only with the driver, but also with the manufacturer, the dealership and even the workshop that performs maintenance.

However, as the article highlights, this definition can raise significant practical concerns, especially considering that Brazil is a country that substantially imports technology. The question is: how to apply this shared responsibility in a market where we are often just consumers of technology developed abroad?

The Challenge of Definition

In my experience supporting startups and companies in their innovation journeys, I have always observed that imprecise definitions are the greatest enemy of effective implementation. The term “AI agents” can quickly become outdated, especially in a field that is evolving at lightning speed.

The legislative urgency, while understandable, may be leading to insufficient exploration of the practical implications of these definitions. It’s like trying to regulate the Internet in 1995—you know it’s important, but you don’t yet fully understand what you’re regulating.

The Enterprise Adoption Paradox

While we discuss regulation, the reality of the Brazilian market reveals a fascinating paradox. 36th edition of the FGVcia Annual Survey shows that generative AI is present in 80% of medium and large Brazilian companies, but is still little used by 75% of them.

This reminds me of the early days of the corporate internet, when companies had websites just to “have a digital presence” without truly understanding the transformative potential of technology. Today, I see the same pattern repeating itself with AI.

Why does this happen? In my experience, three main factors explain this underutilization:

• Lack of team training: Having access to the tool does not mean knowing how to use it effectively.
• Lack of clear strategy: Many companies adopt AI without understanding where it can generate the most value
• Cultural resistance: Changing established processes always encounters organizational resistance

This reality represents a gigantic opportunity for companies that can accelerate their learning curve and effective implementation of AI.

When AI Creates Problems Where It Should Solve Them

One of the most emblematic cases of the last 24 hours was reported by State: AI is creating a big problem for both job seekers and those looking to hire.

The scenario is almost comical if it weren’t tragic: candidates using AI to automatically generate resumes, resulting in more than 1,200 applications for a single job. LinkedIn now registers 11,000 applications every minute, and that’s grown by 45% in the last year.

On the other hand, companies like Chipotle report that using AI chatbots for screening has reduced hiring time by 75%. But this has sparked an arms race: AI generating resumes versus AI screening.

The Lesson of Authenticity

This case perfectly illustrates a principle that I always defend: technology amplifies both human virtues and vices. If we already had problems with mass applications and inefficient selection processes, AI has only escalated these problems.

The solution is not to resist technology, but to redesign processes considering the new reality. Companies that can balance technological efficiency with authentic human assessment will have a significant competitive advantage.

Unexpected Successes: AI in Healthcare

It's not all about challenges. One of the most inspiring pieces of news came from S.Paulo Newspaper: Studies presented in Dublin show that AI-based technologies are significantly increasing the success of smoking cessation programs, with rates of up to 50%.

This demonstrates the power of AI when applied with clear purpose and proper context. The algorithms simulate human conversations, offer personalized encouragement and develop specific strategies to deal with the desire to smoke.

In the UK, the Quit Genius app showed that 45% of users remained smoke-free after six months. In Brazil, the average success rate is 46%, with the potential to reach 70% in specific groups.

This case makes me reflect on something fundamental: AI works best when it complements, not replaces, human expertise. Experts are clear in stating that technology should be an ally, not a substitute for validated treatments that involve psychological support and medication.

What We Learned from Global Challenges

As we face our local challenges, it is worth looking at what is happening globally. news about Apple's problems with AI on the iPhone 17 is emblematic: even tech giants struggle to deliver AI promises on time.

Analysts at JP Morgan point out that delays in AI features, especially related to the new Siri, could hurt iPhone 17 sales. Apple has reduced its production forecast by 9% compared to the iPhone 16.

This teaches us that AI is still a developing technology, and even the world's most advanced companies are learning how to implement it effectively.

The Historic Decision of the STF

A piece of news that went unnoticed by many, but which I consider fundamental, was the analysis of CNN Brazil about the STF's decision to include chatbots in the responsibility of digital platforms.

Carlos Affonso Souza, a specialist in digital law, highlights that this inclusion is significant because chatbots do not qualify as content generated by third parties, but as an action specific to the platforms. This decision sets an important precedent for AI civil liability.

It is interesting to note how Brazil is simultaneously creating specific regulations for AI and setting legal precedents through concrete cases. We are building a robust legal framework that can position us as a global reference.

Reflections on the Future of Work

The article of the Infomoney brings a perspective that, although polarizing, deserves reflection: “AI will take your job. Accept it – and adapt.”

The author makes an interesting historical analogy, comparing the current moment with the transition from rural to industrial society. His message is clear: resistance to change leads to suffering, while proactive adaptation creates opportunities.

I partially agree with this view, but with important nuances. In my experience working with thousands of startups and professionals, I observe that AI is not simply replacing jobs, but redefining the value of human labor.

Jobs that rely on human energy, emotion, and real-time interaction will continue to be in high demand. The key is to integrate AI into the workflow, not resist it.

Opportunities in the Midst of Chaos

Looking at all these news stories together, I see a clear pattern: We are living in a time of transition where opportunities belong to those who can navigate complexity wisely..

For Brazilian companies, this means:

• Empower teams quickly: It’s not enough to have access to AI, you need to know how to use it effectively
• Develop clear strategies: Define where AI can generate the most specific value for your business
• Preparing for the regulatory environment: Understanding the implications of the new legislation
• Investing in hybrid processes: Combining technological efficiency with a human touch

For professionals, opportunities are in:

• Develop skills complementary to AI: Creativity, empathy, critical thinking
• Learning to work with AI: Not against it, but as an amplifier of capabilities
• Focus on authentic human interactions: Where AI still can't compete
• Building intuition about AI: Understanding when to use and when not to use

The Path Ahead

The news of the last 24 hours reveals that Brazil is in a unique position: We are pioneers in regulation, but we are still learning practical implementation. This combination could be our greatest competitive advantage.

While other countries are still debating how to regulate AI, we are building a legal framework. While many still resist adoption, we already have 80% of companies with access to the technology. What is missing is the bridge between these two worlds: effective implementation within a clear regulatory environment.

I see parallels with the mobile revolution in Brazil. We were one of the first countries to massively adopt smartphones, but we were slow to develop a robust ecosystem of apps and services. With AI, we have the chance to do things differently.

In my mentoring work with startups and companies, I realize that Organizations that can balance agile experimentation with ethical responsibility are coming out ahead. They don’t wait for regulations to be perfect, but they also don’t ignore the ethical and legal implications of their choices.

The key is to build internal capabilities while helping to shape the regulatory environment. It is about participating in the game being played, not waiting for the rules to be written by others.

If you are a leader of a company that is one of the 80% that has already adopted AI but is still underutilizing it, now is the time to accelerate. If you are a professional concerned about the future of work, now is the time to empower yourself. If you are an entrepreneur seeing opportunities, now is the time to build solutions that take advantage of our unique regulatory position.

The window of opportunity is open, but it won't stay that way forever. As I always say in my mentoring sessions: the future belongs not to those who foresee changes, but to those who adapt to them quickly and wisely.