Artificial intelligence (AI) is becoming better at producing competent work. Give it a brief, a spreadsheet, a research question or a blank page and, within seconds, it can often produce something that is relatively useful. That is a remarkable productivity gain, because an employee may only need to spend time finessing what the AI has created. However, this may also create a performance problem.
Zynga founder Mark Pincus recently described the risk neatly when he told Fortune that AI can get people to a “B-plus in seconds,” while warning that people may struggle to recognise an A if they have never learned to produce one themselves[1].
For business leaders, that distinction matters. The question is not simply how much faster AI allows people to work. It is what kind of performance an organisation needs from them. There is a difference between algorithmic performance and adaptive performance.
Algorithmic performance is about executing a known process efficiently. Think of replacing the manual tightening of a screw on a production line with a machine. The task is defined and the desired result is clear. If AI can complete that kind of work faster or more consistently, leaders should use it.
Adaptive performance is different. It is the ability to respond when the situation changes, when the answer is unclear or when the old playbook stops working. This is where innovation and strategic advantage are created.[1]
Why elite sport still values practice
Elite sport has lived with technology for years. Athletes can access extraordinary amounts of performance data. Sensors can identify weaknesses. Video analysis can expose tiny technical flaws. Almost every part of preparation can now be measured with extraordinary precision.
Yet access to the technology is rarely the differentiator for long. Rivals can buy the same tools. What separates exceptional performers is what they do with the information. The best athletes use data as an input, not an instruction. They combine it with judgement developed through thousands of hours of focused, deliberate practice. They test adjustments and learn which marginal improvement is worth pursuing.
Over time, they create a gap in skill that becomes extremely difficult to close. Just as importantly, the athlete develops confidence because they understand why something works. They are not simply following an answer provided by a system.
Business leaders should pay attention to that distinction. If employees continually outsource the thinking that sits behind a task, they may still produce acceptable work. What they may lose is the learning that comes from struggling with a problem and discovering why one option is stronger than another. That is how expertise develops.
The danger of removing the difficult bit
This is where the drive for AI productivity can become short sighted. Many of the experiences that feel inefficient are precisely the experiences that build capability. Drafting the first version yourself forces you to organise your thinking. Making a decision without certainty creates accountability.
Getting something wrong, then understanding why, gives you a reference point for the next decision. Remove too much of that process and you risk creating people who can produce answers without building the judgement required to evaluate them.
That may be manageable when the work is routine. It becomes a serious weakness when circumstances change and the organisation faces a problem with no obvious precedent. At that point, a B-plus answer delivered in seconds may not be enough.
Use AI to strengthen the performer
The answer is not to hold AI back. It is to be much more deliberate about where we use it. Leaders should ask whether a task is primarily about execution or development. If the value comes from speed and consistency, automate aggressively. If the value comes from learning how to think, make sure the human still does enough of the work to build that capability.
Leaders also need to keep raising the standard. AI can make average output easier to produce, which means “good enough” will become more common. Competitive advantage will come from people who can see what is missing and keep improving after others have stopped.
That requires a culture where deliberate practice still has a place. People need meaningful feedback and room to make decisions before asking the machine. They also need exposure to exceptional work, so they can understand the distance between competent and genuinely great.
The companies that benefit most from AI will not be those that remove humans from the process wherever possible. They will be those that use technology to reduce low-value effort while protecting the experiences that build judgement.
When conditions keep changing, the bigger advantage is adaptive performance: people who can respond intelligently when the answer is not already known, then find a better one. AI can help us get to competent faster. Leaders still have to make sure their people learn what exceptional looks like.
[1] https://fortune.com/2026/08/20/zynga-founder-mark-pincus-why-ai-wont-save-you/
Adolfo Gomez Sanchez
Adolfo Gomez Sanchez is CEO at GOLD Results. For more than 30 years, Adolfo has studied, tested, and applied the scientific research behind elite performance, drawing on the groundbreaking work of Anders Ericsson, Mihaly Csikszentmihalyi, and Carol Dweck. His work integrates neuroscience, psychology, and performance science to help individuals and organizations unlock higher levels of leadership, growth, and achievement.
