Pattern Recognition — What do we do, what is an AI doing and how shall we decide?
Pattern Recognition — What do we do, what is an AI doing and how shall we decide?
How an overactive mind led me to the discovery of a fascinating phenomon — the one of pattern recognition and its implications in the age of AI.
Yesterday I had a panic attack. I was wondering, why I have it as I haven’t had one for ages. While finding the source of the matter I came across a simple understanding — what if the fact that I put myself out of my normal daily structure, my mind was not able to consistently focus on a task and recognize patterns there, but instead started to doing this pattern recognition with a new “topic” it found instead.
Instead of conneting dots and writing a book with all the different pieces that needed to be connected, it started seeing everything being related in ways that it turned towards the topic of my own health and wellbeing?
“You are tired, your thyroid looks larger than normal, you are feeling sick, you are shaky… oh, and there is this man sitting in his van, doing nothing, what if he is waiting for you to get you and drag you into his van?, and now there is this police car standing close to your apartment door, and your apartment makes creaking, unusual noises, and the door bell that was ringing at midnight clearly wants to send you a message and there will be someone waiting to break into your apartment and getting you. Be prepared.” — full on chaos. And a clear sign that there was a lot of pattern recognition. That led me to having specific anxious and fear-induced emotions and the decision to stay longer awake than needed.
And I was wondering, what is it with that pattern recognition? Where does it come from?
In an article by Robert C. Barkman, PhD in Psychology Today, he refers to our human pattern recognition as a survival mechanism. As mammals we need to connect the dots to finding food, shelter and security. Based on his insights, the neocortex, our outmost layer of the brain and weighing 80% of its total weight is responsible for that form of pattern recognition.
The Cognitive Psychology: A Student’s Handbook by Eysenck, Michael W.; Keane, Mark T. (2003) describes pattern recognition as a cognitive process that matches information from a stimulus with information retrieved from memory.
When we do pattern recognition as human beings, we are often not only seeing and recognizing those patterns, but adding inductive reasoning to it to make sense of them and determine, what our next steps should be given what we found out.
Taking this idea of not only seeing patterns, but also acting upon them a step further, Matt Mattson from the John Hopkins Institute of Neuroscience writes in one of his articles that superior pattern processing (SPP) is the essence and starting point of all forms of the human brain, including our intelligence, imagination, language, beliefs, and invention. And that they can be reinforced by emotional experiences, indoctrination and even psychedelic drugs.
We use patterns consistently and all the time
Now that we got a better understanding of what pattern recognition is, where do we find it? Everywhere. Mammals used it as we have seen for survival. But we also do so. When we are hungry, we look for where we can find food and where it is “safe” to get food. And we connect the dots and various aspects that play a role — sentiments, feelings, hunches, elements from the moment, thoughts, memories from the past, social dynamics of people moving around, specific sounds, noises, smells. It is a complex system that decides — given all these elements and inputs — my decision will be the following. And we can follow that decision to eat at the place that is the best for us now in that moment.
There are many more pattern recognition situations and themes. We look for patterns in investment markets to decide where to invest in, we build imaginary stories based on images we see in front of our eyes self-induced and in the environment by observing and reading and general knowledge input, and we take into account all the information on a road to make split-second decisions whether we move right or left in case there is an accident right in front of our eyes.
Robet C. Barkmans article was written in 2021. One of the sentences in his article is the following one: “Although computer algorithms can spot patterns, an algorithm has not yet been developed that outperforms the human mind”.
Three years later — would we come to the same conclusion?
Would we say: There is not yet an algorithm that outperforms the human mind? Let’s have a look.
Given that there is Artificial Intelligence. Artificial General Intelligence. And a concept I have not yet heard about before — General World Model — introduced by Runway.
When we look at the examples above. Can AI detect whether we should drirve right or left when driving a car and taking in all the input from the outside, other cars, car behaviors? And isn’t a car like a Tesla already driving on autopilot when we allow the system to do it? Such that the car is not only recognizing patterns, but also acting on our behalf? In a way that it is does not need us to make decisions? If we decide to let the computer determine it for us.
And what about investment markets? We make predictions based on what information we have, meta-information and the hunches that we receive. But will we ever have that knowledge of dynamics that a computer has? And what if that computer now is not only seeing and recognizing these patterns, but also deciding upon these patterns and then acting upon them and making those decisions for us? Will it make better decisions than we do?
And what about imaginary stories? if I input a prompt into ChatGPT, as an example the following: “I am out of imaginarly images today and so I ask you, can you create an imaginary story for me.” I get a 9 short paragraph long story about “mystery, adventure, and unexpected friendship” (ChatGPTs words). I add it to the end of this article to read it by yourself. My input was generic, not providing any input images for the AI to build upon. And yet, it was a creative, intelligent story it came up with.
And yet. Can we set the pattern recognition of humans and artificial intelligence systems to the same level? Do they do the job similarly well?
If we look at the amount of data, that is available to both the human mind and the artificial one, we may think that the AI knows more. But does it?
It has more cognitive insight. More data points. More knowledge. It does not know about one or two or maybe 10 or 20 or 100 companies to invest in, but all of them in a specific time-frame. It has superior understanding of the sensories of our own car and that of others, the sensories of the streets and the paths a car may take when it drives irrationally.
But what about emotions, feelings, touch, warmth, bodily sensations, learnings from actions taken, listening, smelling, tasting? Does an AI take these in as well?
When we watch the video of Runway introducing their General World Model, we see that they intent to add sound, smells, words, and images to their models. To enhance creativity. When this will be the case, we can bring down the list of humanly-unique elements down to feelings, touch, warmth, bodily sensations, larnings from actions taken, smelling, tasting. Ah, but wait. What about sensations? Isn’t what a car does the same? It’s car sensations? And what about learnings from actions taken? Isn’t that only visible when we use language to communicate and thus the AI can know that too?
But it is not only about the data, but also about the Pattern Recognition. And for the fact that there are different data inputs, we will see different patterns.
An AI will see an image here, a data point there, a word written and an emotion explained. Given its input, the AI will compute the next best action to take. Hey, have we not seen that before? Yes, when we spoke about pattern recpognition and inductive reasoning. Doing something based on a specific data point like determining when to leave for work given specific traffic patterns or rolling out a new communication campaign based on product usage data. An AI (ML model) may suggest for you to take a specific sales strategy given the numbers from the last season or water your plants as the water will be running out in 3 days.
Is that that new? No. We have seen that already. Models that predict the world. And machines that already do things for us. Planes can fly on autopilot, cars can drive on autopilot. Plants can be watered via a mechanical/electrical system. And sales strategies can be acted upon by i.e. Ava, the Sales Rep Artisan. Machines already make decisions based on data and act upon them without our direct input.
And yet. Are they already superior to us? Given that they do so many things for us already?
What about our pattern recognition of eating something that does not taste good and immediately knowing that we should not eat it as it harms us. What about our sense of smell and knowing we have to run the house as there will be a fire (well, fire alarms could do that for us), what about the demeanor of another person and our hunch whether it feels good spending time with that person. What about patterns interconnected within various domains. What about our legs that give us insights whether we are tired, whether we need to rest? (Or do we measure that with our heart rate and tiredness sensors as well via computer chips?).
You may find out — it isn’t that easily to find those patterns that are uniquely ours. And yet, we are a moving living breathing being. The patterns are seen and felt by us through our innate need to survive. We need them to determine which path to take. We need them to find food, water and shelther. We need them to build bonds and to find out what is supporting us, making us happy, and joyful compared to those things that don’t. We have a purpose for pattern recognition.
Does AI has a purpose? If it does, it has the purpose to support us in that pattern recognition. For now. It is built by us. A tool to enhance our capabilities. We need to get it going. For now. We need to start an airplane and give it an initiating starting point: “we are starting now, you can start your procedures, recognize patterns and act upon them”. We need to start a car and bring it ont the street, so it can start driving. Then the computer can use its power to get all the inputs of the street, make sense of it, recognize patterns, and decide on the next best action. We need to start a computer and a specific AI model and input data, so that the AI is giving us an output, that we can then decide to act upon or not. Or we can allow the AI to act for us, as with AI sellers, AI pilots, AI writers, AI sellers.
But is there an AI, that can already do it all within one system for us? That is our AI seller, AI pilot, AI writer and AI seller in one entity? That when we start our life, it is there? And it is going with us. It is initiated with our birth and never leaves us? and it helps us making decisions based on the patterns it recogizes within us and outside of us?
But what if we run out of battery life? What if we run out of electricity? What if we run out of chips? Will we then run out of us humans too? Because we need those elements to survive as a species altogether? in the form we know it now? What happens if we lose electricity? We lose the AI, but we also lose part of what makes our modern life possible and livable.
There is no answer. But a curiosity. To understand what is possible and what the implications will be. What if for now we decide how to act based on the inputs of the world. But what if we let AI do it for us? First in one area, then in a second, and soon in a third? And what if at one point in time these AI models interact with each other. And they act for us in various interconnected fields. What will happen then? Who controls it then? Or is it an interconnected controlled and chaotic system? Like a huge spiderweb?
Are we then finding ourselves be intrigued not only by our own systems of the workings of the brain and seeing it as a huge spider web (weapon) that is thinking, doing, recognizing patterns but it is on us to make sense of it and only act upon those elements that we find valuable for our goals and objectives, our likes and needs and our fulfillment and proudness? But also by seeing the artificial intelligence models and agents firing, adding, acting upon behalf of those systems being interconnected? And we wonder, what is that for a chaotic, unclear spider web (weapon) running around there?
Then it is not only one system, but two. And when we put them together? Will we find over-firing of it all to create chaos, finding patterns that are not there, finding insights that are biased heavily and acting in ways that are flawed and problematic? Or will we find a way to enhance both systems to make them stronger together?
Whatever it is. It is good, that sometimes, we can simply close the laptop. Make us a cup of tea. Sit down and write an article. Do some sports and look in the sky and feel the warmth of the sun on our arms. We touched, felt, created, and tasted life. Something that — for now — is uniquely ours.
The AIs Imaginary Story that it produced based on my prompt:
In the heart of an ancient forest, where trees whispered secrets of the past and the air buzzed with enchantment, lay a hidden village. This village, unseen by the modern world, was home to the Ailurians, a mysterious race of people who could communicate with animals. The Ailurians lived in harmony with nature, their abilities a well-guarded secret.
Our story begins with Kai, a young Ailurian with an insatiable curiosity about the world beyond the forest. Unlike his peers, Kai’s ability was unique; he could understand and manipulate the elements. This gift was rare, even in a village teeming with magic.
One night, under a blanket of stars, Kai stumbled upon an ancient scroll hidden in the village library. The scroll spoke of a lost city, filled with unimaginable wealth and advanced technology, buried beneath the sands of time. Driven by the lure of adventure and the chance to prove his worth, Kai decided to seek out the lost city.
Kai’s journey led him to cross paths with Lena, a resourceful inventor from the modern world, who found herself stranded in the forest after a failed experiment with a prototype teleportation device. Despite their differences, Kai and Lena formed an unlikely alliance. Lena’s knowledge of technology and Kai’s magical abilities complemented each other, and together they deciphered the clues hidden within the ancient scroll.
Their quest took them through treacherous terrains, from scorching deserts to icy mountains, each obstacle testing their resolve and deepening their bond. Along the way, they encountered creatures of myth and legend, some friend, others foe. Kai’s ability to commune with the elements and Lena’s inventive gadgets turned the tide in their favor time and again.
As they neared their destination, they discovered that the lost city was not just a repository of wealth and knowledge but also a beacon of hope for a world on the brink of ecological collapse. The city’s technology, powered by renewable energy and in harmony with nature, offered solutions that could heal the planet.
However, a shadowy organization, bent on harnessing the city’s technology for their gain, was hot on their trail. In a race against time, Kai and Lena had to outwit their pursuers, unlock the secrets of the lost city, and safeguard its knowledge from falling into the wrong hands.
In the end, Kai and Lena emerged victorious, not just as heroes who had saved the lost city but as guardians of a future where technology and magic worked in harmony to preserve the balance of nature. Their journey taught them the value of friendship, courage, and the importance of bridging the gap between the old and the new.
As they returned to the Ailurian village, Kai with newfound respect and Lena with a wealth of stories to tell her world, they knew that their adventure was just the beginning of a larger story, one where the past and the future could coexist in harmony.