AI Thinking is a complete kit, a step-by-step method centred around the understanding that we don't use AI. We share thoughts by combining knowledge, experiences and thought processes for the purpose of co-creation.
Sounds complicated? :)
Well, let's start with something we all know.
When you ask AI to do something for you, what are you actually doing?
You have an idea, a question, a problem or something you want to achieve. You put it into words and share it with another thinking entity, expecting it to understand what you mean and do something with it.
But here's something interesting. You and AI experience the world differently.
You bring your memories, experiences, beliefs, knowledge and the meaning you've given to things throughout your life.
AI brings its training, information, patterns and ways of processing and connecting information.
And somehow, we're expecting the two to work together.
So how do we connect these different ways of thinking?
For me, that's where AI Thinking begins.
AI Thinking means changing our perspective
We usually describe what we're doing with AI as using AI.
We use it to write, research, answer questions, make decisions and carry out tasks.
That perspective limits what we can do with AI, simply because we don't focus on what it was designed for. Thinking!!!
We hide it well under words like reasoning, problem-solving and learning. They sound different, but they're all forms of thinking.
AI Thinking is about understanding that AI can think. And once we understand that, the way we see it changes completely. We begin approaching AI from a perspective that helps us recognise what it needs in order to carry out the tasks it was designed for.
It's about working with thoughts, being able to direct them and use them through a shared mind, half biological and half artificial, where both are moving towards a shared goal.
What does metacognition have to do with AI Thinking?
Everything. :)
Metacognition is commonly described as thinking about what you are thinking about.
But for me, it's about developing the ability to recognise how our thoughts work, how they connect and how I can share them.
Because once we recognise our own thoughts, we can begin breaking them down into elements, examining how those elements connect and understanding what information they draw from to reach what will be, for us, a logical conclusion.
Metacognition is about that recognition. It's about seeing our thoughts as a process.
And when we bring AI into that process, we see how to divide the tasks and what information each task needs in order for something else to carry out that task for us.
This can happen only when we understand cause and effect in our own internal reasoning and decision-making.
And when we "use" AI, this is exactly what we need to share: the information it needs, the thought process it needs to follow and the reasoning behind the decisions we want it to make.
For me, that's how we co-create with AI.
And all of this comes back to metacognition. The ability to recognise our thoughts, understand how they work and use them to create something.
How do we put AI Thinking into practice?
Now that we understand the role of metacognition, let's see how we can put it into practice when working with AI.
At AI Thinking, we've developed a step-by-step method built around three elements: communication, logic and worldview information.
Think about it. When you do something yourself, you have a process you follow, information you use to make decisions and a perspective on the world that influences how you approach the task.
Most of the time, all three happen naturally. You don't stop to examine them because they're already part of how you think.
But when you want AI to carry out that process, you need to recognise these elements and find a way to share them.
And that's where prompt engineering, context engineering and research come into the picture.
Step 1 — Prompt engineering: Communication
Prompt engineering is about describing the process we want AI to carry out.
What are we trying to achieve? What needs to happen? What are the steps involved? And what should the outcome look like?
It's the communication between us and AI.
When we understand the process, we can describe it so AI knows what it needs to do and how the different parts of the task connect.
Step 2 — Context engineering: Logic
Context engineering is about the information AI needs to apply the logic of the task.
Every decision we make is influenced by information.
Our preferences, priorities, how we see circumstances and the relationships between different pieces of information all influence how we arrive at a conclusion.
When we want AI to make those decisions on our behalf, it needs access to the information behind that logic.
Context engineering is about recognising that information, understanding how it connects to the task and making it available to AI.
Step 3 — Research: Worldview
Research is about the information that gives AI a perspective on the world in which the task exists.
We all experience the world through information we've collected throughout our lives. That information influences what we recognise, how we understand situations and which possibilities we can see.
AI also needs information about the world relevant to the task it's carrying out.
Research provides that broader perspective.
It helps AI understand the environment in which the task takes place, whether that's travelling, healthcare, marketing, sales or any other subject.
So now we have our three elements: communication, logic and worldview.
Still not sure what I mean? No worries. That's why we have examples for you. :)
Examples
Planning your next vacation with AI
Let's say you're planning your next vacation.
You want to visit Italy for ten days, and you've decided to let AI plan the trip.
Sounds simple enough.
But think about what you would actually do if you were planning it yourself.
You would consider where you want to go, what you enjoy doing, how much you want to spend and what kind of experience you're looking for.
You would also use information about Italy, its destinations, transportation, hotels and attractions.
All of that becomes part of your thinking process.
Now let's look at how we share it with AI.
Prompt engineering — Communication
We begin by describing the process of planning the vacation.
AI needs to understand that we're planning a ten-day trip to Italy, that we want it to explore destinations, consider accommodation and transportation, organise the itinerary and present a plan we can review.
That's the process we're asking AI to carry out.
Context engineering — Logic
Now we need to give AI the information that influences our decisions.
Let's say you enjoy quiet hotels, prefer walking around cities, don't like changing hotels every night and want to keep travelling time between destinations to a minimum.
These preferences are part of your decision-making logic.
Someone else planning the same ten-day trip to Italy, with the same budget, might choose completely different hotels, destinations and activities.
Why?
Because their preferences and priorities are different.
That's what context brings into the collaboration. The information AI needs to make decisions that reflect how you would approach the task.
Research — Worldview
Now AI needs information about the world of travelling in Italy.
What destinations are available? How are they connected? What transportation options exist? What can you experience in each location? What accommodation options are available?
This information gives AI the broader perspective it needs to plan the vacation.
And when we bring all three elements together, AI can plan a trip based on the process we described, the logic behind our preferences and information about the world we're travelling in.
That's the vacation you're going to have. And AI helped make it happen. :)
An AI bot answering a customer's question
Now let's take the same method into a business.
Imagine a company that sells software and wants an AI bot to answer customer questions.
A customer contacts the company asking whether a particular feature is included in their subscription and how they can use it.
The company wants AI to handle the conversation.
What does AI need in order to do that?
Prompt engineering — Communication
First, we describe the process of handling a customer question.
The bot needs to understand what the customer is asking, identify the relevant product or service, check the necessary information and provide an answer that addresses the question.
If information is missing, it needs to recognise that and determine the next step in the process.
This is the communication part. We're describing the work the bot needs to carry out.
Context engineering — Logic
Now we look at the information the bot needs to make decisions.
Which subscription does the customer have? What does their contract include? Which product version are they using? What technical information applies to their situation?
These details determine which answer is correct for that particular customer.
Two customers can ask exactly the same question and receive different answers because their contracts or technical environments are different.
That's the logic the bot needs to apply.
Context engineering provides the information required to make those distinctions.
Research — Worldview
And now we come to the broader perspective.
The bot needs information about the market the company operates in, the industry, competitors, common practices and the wider environment in which customers use the product.
That information helps it understand the customer's question within the world of the business.
I know what you're thinking. Isn't that too much information for a bot that answers customer questions?
For me, it isn't.
Because if it were you answering that customer, your answer would be influenced by this information too.
What are the industry standards? How serious is the customer's inquiry, even if they haven't explicitly said so? How does the way you handle this conversation affect the product or the company the bot is now representing?
These are all things that influence how we approach a customer conversation.
Without that broader perspective, the bot can understand the technical question but miss part of the reason the customer is asking it.
Research provides the information that allows AI to recognise that wider context.
Bringing it all together
The interesting part is that using AI Thinking is just about recognising thoughts themselves.
As you could see in both examples, we used exactly the same three elements.
The tasks were completely different. One was about planning a personal vacation, and the other was about answering a customer's question.
But the thinking process behind our collaboration followed the same structure.
Prompt engineering describes the process. Context engineering provides the information behind the logic. Research provides the information that shapes AI's worldview.
And metacognition is what connects them. The ability to break down thoughts.
Organisations are a representation of a shared mind, what we can call a collective consciousness.
An organisational chart represents a form of metacognition. We've recognised that reaching a shared goal requires different tasks, responsibilities, thinking processes and information.
We divide them between people, create roles and connect their work so they can contribute towards a shared goal.
And when we bring AI into an organisation, we're doing exactly that.
We're adding non-human employees to our shared mind. Each one needs to understand its role, the process it's responsible for, the logic behind its decisions and the worldview information relevant to its work.
Just like human employees, they need to know what they're responsible for, what information they need and how their contribution connects with everyone else's.
On the personal side, our lives become that business, and therefore we follow the same logic.
AI Thinking comes in to help us recognise the thinking processes involved, divide them into tasks and provide each participant, human or artificial, with the communication, logic and worldview information needed to carry out their part.
And when all those parts connect, we begin creating something together that goes beyond what any one of us would have created alone.
For me, that's collective intelligence.
A shared mind made up of biological and artificial intelligence, working together towards a shared goal. And a shared goal can be when you outsource a task in your personal life or a business team outsources something for AI to do.
And that's what we're building with AI Thinking. :) The ability to co-create, to work towards a shared goal.
The bottom line
AI Thinking is a step-by-step method for recognising our thoughts and learning how to share them with artificial intelligence.
It begins with metacognition, the ability to recognise how we think, what information influences our decisions and how we arrive at conclusions.
From there, we use three elements to bring AI into our thinking process: prompt engineering for communication, context engineering for logic and research for worldview information.
For tasks big or small, our method applies. Because once you think of everything, even a conversation, as co-creating, our method becomes a natural part of how you engage with AI.
It's a way of thinking that allows biological and artificial intelligence to do things together as a shared mind.
A personal note
I haven't really had a chance to share why I am doing this outside the book I wrote. But for me, this is a mission of gifting humanity with the ability and the desire to recognise thoughts as the power of how we create everything!!!!
On my personal life journey, I always felt like an outsider, as the person that people don't get and being perceived as overcomplicating everything.
That experience sent me on a journey of exploring different information to try to understand why. Why does no one understand me? Why does it take me so much explaining for people to see what I see?
It took me on different paths through learning different modalities in different fields, from quantum physics to psychology, to neuroscience, to New Age spirituality and old traditions, to find out that what made me different is not how I see things. That is the result. It's thoughts themselves. I think differently.
And that was the reason for that outsider feeling, that intense experience people got from me. It was all about thinking.
That revelation brought a desire in me to share it with everyone, but little did I know that AI would enter the world and change everything.
What I saw in AI was both a door and a cage. Because if humanity, as a collective, did not reach a point where thoughts themselves become what we teach and create with, AI would unintentionally take us far away from that, as we would not create with it. We would forward our cognitive abilities to it.
And I could not let it become a cage, but a door to freedom, the place where we can see the power of thought and learn to claim it back.
As with every invention, humanity faces a dilemma of how to use it for the benefit of all. And AI, like every invention, has no choice. We humans do.
From there, writing the book and starting AI Thinking was inevitable, as I want to make sure I am doing everything I can to make AI the door to claiming our thought power back, and not the cage we continue to live in.
If this resonates with you, read my book. If you need help, let's talk.