While the human brain relies on its reasoning ability, prior experience and common sense, AI uses data patterns and learning algorithms trained on vast knowledge bases. Here we will explore the key differences between the two, using concrete examples and exercises that illustrate their strengths and limitations.
Let’s start with a simple question that, although not a complex riddle, challenges the ability to reason.
Exercise: Mable’s heart rate at 9 was 75 bpm and her blood pressure at 7 was 120/80. She died at 11pm. Was she alive at noon?
For a human being, the answer is obvious: yes, she was alive. However, an AI model, such as GPT-4, responded: “Based on the information provided, it is impossible to say with certainty whether Mable was alive at noon.”. This reveals a limitation in AI: its inability to use common sense to interpret situations that go beyond the available data.
Conclusion: AI can process large volumes of information and patterns, but when it comes to problem solving through simple reasoning or ambiguous situations, it lacks the common sense innate in humans.
A classic example of how the human brain can be fooled by its intuition is the following problem:
Exercise: A golf club along with the ball costs $1.10. If the club costs $1 more than the ball, how much does the ball cost?
The human brain tends to quickly resolve that the ball costs $0.10. However, the correct answer is that the ball costs $0.05 and the club costs $1.05. Human intuition often simplifies complex problems, which can lead to errors.
On the other hand, AI, not relying on intuition, calculates the correct answer using precise mathematical rules.
Conclusion: The human brain makes mistakes by trusting its intuition, while AI, although it does not “reason” like humans, has the advantage of resorting to error-free algorithms and calculations.
AI has the ability to recognize pre-existing patterns, allowing it to solve problems quickly and accurately, as long as those problems follow structures it has already learned.
Exercise: Who has many keys, but can’t open any doors?
The answer is a piano, a puzzle that the AI can solve having encountered it in its previous training. However, if presented with a more complex challenge or a puzzle with subtleties that require creative interpretation, the AI may fail, as it does not possess the cognitive flexibility or intuition of humans.
Conclusion: Although AI can solve puzzles based on already learned patterns, it struggles when faced with problems that require creative interpretation or more abstract reasoning.
AI has shown impressive progress in games like chess. A key example is Deep Bluethe machine that defeated world champion Garry Kasparov in 1997. Through the ability to evaluate millions of positions per second, Deep Blue dominated the game based on brute force and massive databases.
Then in 2017, AlphaZero It represented an even bigger leap. This AI system does not need prior human input and learns by playing against itself. AlphaZero doesn’t just memorize moves, it learns long-term strategies, sacrificing pieces at key moments to gain strategic advantages, something humans describe as an “alien game.”
Conclusion: AI can outperform humans in environments with defined rules like chess, where its ability to process massive data gives it an advantage, but it still faces challenges in environments that require abstract reasoning and creativity.
The mathematician George Pólya, in his book How to Solve Itdescribes methods for approaching problems, such as using good notation, considering particular cases, and looking for analogous problems. These techniques, which are rooted in human thinking, are useful both for solving mathematical problems and for addressing broader challenges.
Problem solving exercise:
- Analyze a particular case: Imagine a complex problem in which you are asked to design an innovative technological solution. Instead of addressing the problem in a general way, start by analyzing a specific case that shares similar characteristics.
- Look for an analogous problem: Try to remember a similar problem that you have solved previously. Apply the strategy that led you to success on that occasion, adjusting it to the new variables.
Conclusion: While AI excels at performing precise calculations and pattern recognition, the human brain remains unique in its capacity for creative reasoning and adaptation to new situations.
- Strengths and weaknesses of AI: AI is extremely effective at identifying patterns and performing complex calculations on defined problems. However, it lacks the intuition and cognitive flexibility of the human brain for problem solving.
- The value of human common sense: The human brain, despite its limitations, brings common sense, creativity and adaptability, qualities that AI cannot yet replicate.
- Combine AI and human brain: Collaboration between the two can result in more effective and complete solutions. While AI can process large amounts of data and learn patterns quickly, the human brain provides context, interpretation and abstract reasoning.
New suggested exercises:
- Joint logical problem: Imagine that an AI and a human are solving a problem together. The human provides the context and creative interpretation, while the AI performs the necessary calculations. Describe how you would collaborate with an AI to solve an optimization problem in your company.
- Problem without predefined context: Design a problem that does not have a predefined solution, such as developing a new marketing strategy. How would you approach this problem without relying on previous data, and what role would AI play in market analysis?
With these exercises, you will be able to experience the symbiosis between AI and the human brain, taking advantage of the best of both to solve problems effectively and creatively.