Turn your adversary into a friend
The AI cycle continues...
Si quieres leer o compartir este artículo en español.
Image development and text assisted by ChatGPT
I still remember when UBER arrived in Barcelona and the incredible uproar it caused among taxi drivers. They had paid over 100,000 euros for licenses to operate in the city, and suddenly, new competitors appeared—drivers who, by simply downloading an app, could start picking up passengers and earning money. The taxi drivers began organizing boycotts, blocking roads, and even attacking the vehicles and drivers of this "competition." They failed to see that they were "shooting bullets at a dead horse"… the rental of chauffeur-driven vehicles was here to stay.
Since the second phase of industrialization began in 1850, technology and innovation have continuously disrupted how we interact, work, and live our daily lives. You’d think that after almost two centuries, we would have understood this by now… but no. We still throw tantrums like children, demanding that the "daddy-government" to step in and regulate everything. It’s as if we have shifted our complaints from God to the government.
Those who seem like adversaries remind us that we must grow and change. As long as we continue behaving like children—expecting others to restore order—we will keep having governments that treat us as such and societies that follow blindly like sheep.
I will repeat this a thousand times: You have freedom of thought and expression so that you can shape your own life without restrictions—not so that others can protect you from the hard realities of life.
Let’s make an effort to understand how this formidable AI tool works. It mimics just one simple function of our brain, yet we have so many more capabilities at our disposal. Instead of resisting, let’s use our full potential to navigate this "perfect storm"—which is nothing more than an amazing opportunity to grow and develop ourselves.
It all starts with knowing the adversary… then we can figure out how to turn it into an ally.
AI is designed to help us answer our biggest questions. So, how can we make the most of it?
Large Language Models (LLMs), like those used in ChatGPT and other AI tools, function as a "massive problem-solving engine" for almost any prompt we provide. When I first embarked on this journey, I evaluated various models to determine which was the most competent and versatile. I ultimately chose ChatGPT, but you should use whichever one best suits you.
The following tips apply to all LLM tools:
First: CREATE YOUR ACCOUNT AND LOG IN. There are benefits to registering.
Second: STAY ORGANIZED. Your chats will be organized based on the first question you ask. Keep one topic per chat, as the system remembers your requests. If you provide instructions on how to translate a text, for example, you won’t have to rewrite them every time. You can also edit the chat title for easy reference.
Third: LEARN TO ASK QUESTIONS. This is the key to unlocking the full potential of the tool—or, for that matter, to living a fulfilling life.
Learning to ask the right questions helps you discern the correct path forward.
Throughout my articles in this series, I have emphasized the need for growth and development to face this formidable adversary. The ability to ask insightful questions is a skill that will yield immense benefits in your life. The paradox is that if you don’t have a clear goal, any direction is a good one.
I encourage you to read my article Hope: A Superpower! where I explore why having a meaningful goal is crucial and the benefits it brings.
AI is an powerful ally in clarifying your path and guiding you toward your objectives. I will now outline the five key areas you should focus on to obtain effective responses:
Area: Clarity
Objective: Obtaining precise and useful answers depends on how clear and specific you are.
Detail or Example: Specify key details such as purpose, context, and constraints.
Area: Context
Objective: Personalizing the response according to your needs—what are the background details? What is the specific situation?
Detail or Example: Allows responses tailored to the user’s specific situation. Provide information on the problem, background, and constraints.
Area: Response Format
Objective: Receiving information in the desired format.
Detail or Example: Indicate if you prefer a bullet-point list, a table, an example, programation code, a summary, etc. Define and clarify how you want it presented.
Area: Depth Level
Objective: Adjusting the complexity of the response.
Detail or Example: Allows responses to be more technical or simplified depending on the user’s level. Do you want a story that a 9-year-old can understand or a technical explanation of how “X” works?
Area: Purpose of the Response
Objective: Adapting the response to a specific goal.
Detail or Example: Clarify whether you need the information to learn, make decisions, generate ideas, solve a problem, etc.
I'll give you the example I used to test ChatGPT vs DeepSeek when all the hype about the launch of the new Chinese APP. I wanted to ask a simple question but one that had a complex analysis process to answer. I also wanted to ask a question that was close to its reality: the history of AI [Ref.: article ChatGPT VS DeepSeek].
My question was:
Can you please give me a brief history of AI?
Include a table of the most significant milestones in history.
Include the year and why they were significant.
Also include the most relevant people in that history.
It was the simplest question with constraints that challenged the LLM to show me if he had processed the information relevant to its own story.
If I had wanted to be more demanding I could have asked the following question:
Can you please give me a brief history of AI [clarity]
I need it to teach 4th graders [context]
I want you to write a 1500 word short story to read in class [answer format and depth level]
I also want you to give me a table for me to use as a reference [context and format] where you include the following columns: year, milestone in AI history, relevant people involved in it, how it advanced AI history… important to begin from ‘neural networks’ when it was first invented [answer format and depth level]
As you can see, this answer is infinitely more complex at multiple levels. Just now I asked the question to both of them and I would like to share with you the introduction to each of their stories:
DeepSeek Story: "The Dream of Intelligent Machines"
Once upon a time, in a world not unlike our own, there was a group of dreamer-scientists who wanted to create something never seen before: machines that could think like humans. This dream did not come about overnight, but rather grew little by little, like a tree that spreads its roots and branches over time.
[DeepSeek wrote the story first and then the table.
ChatGPT Story : (no title)
Once upon a time, in a world very similar to ours, there lived a small town where three very curious friends lived: Sofia, Leo and Max. They loved learning about the universe, dinosaurs and robots. But what intrigued them most was something their teacher mentioned in class one day: "Artificial Intelligence.""What is Artificial Intelligence?" Sofia asked, her eyes shining with excitement.
[ChatGPT wrote the story and the table at the same time creating a screen division]
I trust you recognize the difference between DeepSeek and ChatGPT… but I’m moving on now because we still have something important to cover…
Formats and areas of response and solutions
I know this part can be tedious, but I urge you to take some time to understand how to generate a “Command” (English: prompt). Think of it like you are training a ‘personal assistant’ to give you its very best… but, it is not AI that you are training, it is yourself, you need to understand all the areas you can ask for support and how to do it.
Area: Specific restrictions or limitations
Objective: Helps define what you want when you eliminate or constraint what you don't want.
Details or examples: One of the most effective tools to delve deeper into the objective is when you set limits on what must be answered. Also when you ask that certain elements be marked: "review this text and correct spelling and syntax; mark all corrections you make in bold."
Area: Comparisons and Evaluations
Objective: Obtain advantages, disadvantages and best options.
Details or examples: Ask for comparisons between options, pros and cons or specific recommendations. It facilitates informed decision making.
Area: Use Cases and Applications
Objective: Understand how to apply the information in practice.
Details or examples: Ask for specific examples, use cases or real situations where the information is applied. It makes the response more useful and actionable.
Area: Content Generation
Objective: Create structured texts, ideas or materials.
Details or examples: Specify the type of content (article, script, report, story, etc.). Save time on writing and creativity.
Area: Clarify tone or style
Objective: Defining the tone gives it a touch of style and form.
Details or examples: You can ask it to write a text as a technical specification; or an email with a friendly and approachable style; or give it a text of a story and ask it to write it as a horror novel.
Area: Problem Solving
Objective: Obtain action-oriented responses.
Details or examples: Explain the problem in detail and request steps, strategies or approaches. Facilitate finding concrete solutions to specific challenges.
Area: Optimization, Improvement and/or Translation
Objective: Refine ideas, texts or strategies. Perform a translation.
Details or examples: Ask for revision, correction or suggestions for improvement for a text or strategy and/or request a translation. It allows you to improve the content, structure or impact of a proposal.
Area: Validation and Feedback
Objective: Confirm whether an idea or approach is solid.
Details or examples: Requests critical evaluation or alternatives to an idea, argument or decision. Reduces errors and improves decision making.
Area: Automation and Code
Objective: Obtain code or solve programming problems.
Details or examples: Indicates the programming language, the problem and any technical restrictions. Speeds up development and error resolution.
Area: Strategy and Planning
Objective: Obtain structured approaches to achieve an objective.
Details or examples: Requests strategies, roadmaps, checklists or methodologies adapted to a purpose. Helps to better organize steps and priorities.
Area: Persuasion and Communication
Objective: Improve the argumentation and effectiveness of the message.
Details or examples: Requests examples of persuasion, communication techniques or improvements to a message. It allows you to transmit ideas in a clearer and more impactful way.
Imagine that you have at your disposal a person capable of helping you with any goal or objective you may have, but in order for them to truly give you all the potential they have, you need to be very clear about how to tell them what you want. You have to grow and learn so that you can get the most out of this “plenipotentiary assistant”… but their help will not be better than what you can discern or take advantage of.
And well… like everything in life, nothing good is exempt from its pitfalls; having ‘someone’ so powerful at your disposal can make you, by relying on them so much, not develop the character and temperance you need to manage that power… do not let yourself get drunk with the sweetness of its offer, again remember Frankenstein or the Tower of Babel.
Next week…
There is no shortcut to greatness
To make AI a friend, you have to grow and elevate yourself and stop beingnot become a beggar.
