Artificial intelligence is real and its impact is real. But the market of vendors, consultants, and candidates who call themselves AI leaders has a structural problem: nobody is one yet. This is a guide for leaders who must decide amid the noise.
Don't be dazzled by a vendor who claims to be an AI leader. Nobody masters AI, nobody is an AI expert. We are all amateurs learning. The difference lies in doing it consistently.
If in the last six months you haven't been sold at least one artificial intelligence solution, a digital transformation consultancy with AI at its core, or a candidate presenting themselves as an 'AI expert'... it's because you haven't attended enough business meetings.
Artificial intelligence is real. Its impact is real. But the market of vendors, consultants, and candidates who call themselves AI leaders has a structural problem: nobody is one yet.
According to EY data for Colombia, 92% of workers already use some AI tool in their daily work. However, only 28% of organizations globally have managed to turn that adoption into real business transformation. That gap is not technological. It is a gap in executive judgment.
This article is for business leaders, human resources directors, and board members who have to make decisions amid this noise. We are going to tell you, plainly, how not to be fooled and how to build real AI capabilities in your organization.
1. The problem is not AI. It is the 'experts' narrative
There is an interesting paradox in the artificial intelligence market today: the technology advances faster than the human capacity to master it. This means that anyone who proclaims themselves an AI 'expert' is, at best, being optimistic; and at worst, selling you smoke.
The world's most advanced language models —GPT, Claude, Gemini, Llama— released new versions in the last twelve months. Agentic AI implementation frameworks change every quarter. AI governance best practices are being written right now. Not two years ago. Now.
In this context, the concept of 'expert' loses meaning. Not because there aren't brilliant people working in AI —there are—, but because the field moves so fast that twelve-month-old experience can be irrelevant today.
What does exist, and what you should look for, is something very different from expertise: consistency in learning and practical application.
Key fact: Corporate enrollments in generative AI courses in Colombia grew 288% year-over-year in 2025–2026, according to Coursera. The market is learning in real time. Anyone who tells you they have already reached the destination is lying to you.
2. The warning signs: how to identify a vendor that is 'dazzling' you
At Jugada Maestra we have seen how large Colombian and Latin American companies hire AI vendors based on leadership narratives that do not withstand the slightest scrutiny. These are the signs that should set off your alarms:
Sign 1: They use the word 'leader' without concrete evidence
'We are AI leaders for the financial sector.' Leaders according to what metric? What projects have they implemented? What was the measurable impact on P&L? If there are no clear answers to these three questions, there is a problem.
Sign 2: They don't talk about failures or limitations
According to AI Summit Colombia 2026, only 5% of AI projects are fully implemented. A vendor who doesn't tell you about their failed projects, what they learned, and how they adjusted does not have the maturity your organization needs.
Sign 3: They propose generic solutions for specific problems
'We implement AI across your entire value chain.' That phrase means nothing. AI works when it solves a concrete, measurable problem, with available data and a clear business case. Be wary of anyone who cannot articulate that from the first conversation.
Sign 4: They lack clarity about their own models and tools
Are they building with proprietary models or using third-party APIs? What is their position on data privacy and security? How do they guarantee their solution will remain relevant in twelve months? If they don't have clear answers, you are paying for a thin presentation layer on top of someone else's technology.
Sign 5: They promise speed without talking about governance
Implementing fast without structuring governance is a recipe for chaos. 52% of Colombian companies already acknowledge that technological fragmentation prevents them from making informed decisions, according to SAP. A responsible vendor talks about integration, roles, data, and control from day one.
3. The profile you actually need: the consistent learner, not the guru
When a company asks us how to search for executive talent with AI capabilities, our answer usually surprises them: don't look for an expert. Look for a consistent learner with good judgment.
The difference lies in doing it consistently. Not in knowing it all. Artificial intelligence does not reward the one with the most theory —it rewards the one who has experimented, failed, adjusted, and tried again the most.
In our experience mapping talent for senior management in Colombia, Panama, and the rest of Latin America, the executives who are generating real value with AI have three characteristics in common:
- Clarity about what AI can and cannot do in their specific business context
- A history of continuous experimentation: pilot projects, documented failures, real iterations
- The ability to translate technology into business language for their teams and their boards
What they don't have: 'AI expert' certifications from vendors that have been in the market for three years.
Are you building or strengthening an executive position with AI responsibility in your company? At Jugada Maestra we help you define the right profile —not the fashionable one— and find the talent that truly generates impact.
4. How to evaluate AI vendors and candidates with executive judgment
Here are the questions you should ask —and that your counterparts should be able to answer fluently— before making any AI-related decision:
For AI technology or consulting vendors:
- Can you show me a project that failed and what you learned from it?
- How do you measure the success of your implementations at 12 and 24 months?
- What happens to my data? Who controls it, where is it stored, how is it protected?
- Which AI model do you use and why that one and not another?
- How many projects similar to mine have you fully implemented —not just started—?
For executive candidates with AI responsibility:
- Tell me about an AI project you led. What worked, what didn't, and how did you adjust?
- How do you explain to a board of directors when it is worth investing in AI and when it is not?
- What tools do you personally use today? How has your use evolved over the last year?
- What is your position on the ethical and bias risks in the models your organization uses?
- How would you build internal AI capabilities in a team starting from scratch?
A candidate or vendor who answers these questions honestly —including the parts where they don't yet have all the answers— is infinitely more valuable than one who has a perfect answer for everything.
5. The strategy that does work: building with consistency
The most important conclusion of this article is also the most uncomfortable one: there are no shortcuts. The only sustainable competitive advantage in artificial intelligence is building your own capabilities in a constant, systematic, and honest way.
What does building with consistency mean in practice?
- Dedicating regular time —not one-off projects— to learning and experimenting with AI in your area of responsibility
- Establishing clear metrics before implementing any AI solution, not after
- Creating a culture where mistakes in AI implementation are documented and learned from, not hidden
- Prioritizing data governance as a precondition for any AI project —without clean, well-structured data, no AI works—
- Evaluating your vendors and collaborators by their capacity for continuous learning, not by their certifications
66% of AI projects in Colombia have already passed the initial integration phase, but only 26% have reached an advanced stage of real use. The difference between those two groups is not technological. It is one of discipline, judgment, and consistency.
Frequently asked questions about AI and executive decisions
Is there anyone who really is an AI expert?
There are indeed people with deep knowledge in specific subareas: machine learning engineering, model architecture, AI ethics, implementation in specific sectors. What does not exist is the generalist 'AI expert' who masters the entire field. When someone presents themselves that way, it is a warning sign.
How much should an AI implementation consultancy cost for a mid-sized company?
Ranges vary enormously depending on scope. One thing is clear: if the price is not tied to measurable results and specific deliverables, it is a sign of trouble. Always demand a model where the vendor has skin in the game.
How do I know if my company is ready to implement AI?
The right question is not whether it is ready, but for which specific use case it is ready. Most companies have at least one process where AI can generate value today, without needing a complete digital transformation.
Is the AI talent market overvalued in Colombia?
Yes and no. There are very specific technical profiles where demand genuinely exceeds supply. But there is also title inflation, where people with basic knowledge of AI tools present themselves as specialists. The key is to evaluate concrete projects, not credentials.
When does it make sense to hire a Chief AI Officer or an AI Director?
When your company has AI projects in multiple areas that need strategic coordination, when data governance and AI risks have become a real concern for the board, and when you already have accumulated in-house experience —not as a first step—.
Conclusion: The advantage lies in honesty
At Jugada Maestra we have spent more than a decade helping companies in Colombia, Panama, and Latin America find and evaluate executive talent. And what we have learned about artificial intelligence is exactly the same thing we have learned about leadership in general: the most dangerous person in any organization is the one who pretends to know what they do not know.
Artificial intelligence has no owner yet. Nobody masters it. And that, well understood, is not a weakness of the market —it is an opportunity for organizations that have the humility to learn seriously, the discipline to do it consistently, and the judgment not to be dazzled by empty narratives.
The difference is not in who claims to be a leader. It is in who keeps learning when nobody is watching.
Are you building an AI-related executive position, or evaluating vendors for a strategic initiative? At Jugada Maestra we ask the hard questions before you have to.