It's not a very precise question. because what does it really need to learn and like a human I mean how close do you need to be to a human for that to be true, but in the short answer: no! It's because they don't work as humans do. There are so many things we don't understand about the human brain and how it works and even though we're inspired by it when we build AI does mean that it's doing the same thing it's definitely not doing the same thing. One example would be large language models such as chat GPT that the way it produces answers is that it's trying to predict answers, it's guessing 1 word at a time which is very different from at least part of what makes up an answer when a human answers the question. Of course part of what I'm doing right now is that I might not actually know where my sentences are going to end but I know before I started answering I knew what I wanted to say I know the kind of meaning I want to convey I'm thinking about you as a like a receiver and how you might perceive my answer so I'm doing a lot of things that as fast we know are not part of these language models said at least they were not never explicitly programmed to have these abilities so and that quite beheaded planning. Then we have other types of AI that are good at planning but bad at learning so I would say the answers definitely no to your question that they don’t learn and think like humans and it’s simply because like in a way why would they? Okay they impress and then you start thinking that they maybe do, right? But what they do and how their program is still fundamentally different, so it's not to be expected I would say. Maybe in a way it's even more impressive when they are human-like in their let's say answers or their behaviour then when they are not.
No. Today’s AI neither thinks nor learns in the same way humans do. The differences between human cognition and artificial intelligence are fundamental. AI systems identify statistical patterns in data, whereas humans construct meaning and genuinely understand relationships between ideas, experiences, and contexts. AI operates by optimizing predefined objective functions. Humans, by contrast, are able to question, reinterpret, and reflect on goals themselves. We can change our intentions, evaluate values, and consider ethical or emotional dimensions when making decisions. AI lacks all of these capacities. It has no consciousness, no intentionality, and no real understanding of the content it processes. What is commonly referred to as “learning” in AI is, in reality, the mathematical adjustment of parameters within a model based on training data. This process does not involve comprehension or insight. It is an optimization procedure, not an act of understanding. Similarly, the term “thinking” is used metaphorically rather than functionally. AI does not reason in the human sense; it produces outputs based on probability distributions derived from past data. Furthermore, AI’s ability to generalize is limited to the data and structures it was trained on. Outside of this space, it lacks the flexibility and contextual awareness that characterize human intelligence.
In short: No – today's AI does not learn and think like humans, even if it sometimes seems that way. Why does it still seem that way? Modern AI (especially with neural networks) can evaluate huge amounts of data and recognize patterns in it. As a result, she can use language, understand images or make decisions surprisingly well. This often feels "intelligent". The key difference: People understand things, have awareness, goals and experiences. AI optimizes patterns – it calculates probabilities for what "fits". A simple image: A person learns what a "dog" is by experiencing the world, understanding meaning and forming concepts. An AI learns "dog" as a statistical pattern in data (pixels, words, etc.), without any real understanding.
This question has been floating around for many years, with the scientific community trying to resolve it, but there is no clear or definitive answer. One of the most famous approaches to this question is that of the Chinese Room. Imagine a room with two slits: one through which sheets of paper with questions or written instructions are inserted, and another through which the answers come out. The instructions inserted are always in Chinese. Inside the room is a person who has no contact with the outside world other than through the two slits. This person knows nothing about Chinese but has a bunch of manuals that tell them what to return through the other slit depending on the characters they receive. So, to any Chinese speaker who wants to communicate with the room, it will seem that the room (or whatever is inside it) knows Chinese, because it responds correctly to every instruction or question, they give it. However, as we have already said, nothing and no one in the room knows the language, no matter how much it may seem that way. Using this example, it is argued that AI will never think like humans, since nothing within it really resembles what we would call “thinking”; it only imitates our behaviour (for now). However, returning to the previous example: nothing within us knows English or Spanish; not our stomach, not our spleen, not our neurons individually. But we, as a whole, do know languages. Seen in this light, we are not so different from AI. So why can we think and they cannot?
This is more of a philosophical question, but I think it is not capable of truly learning or thinking, and it won’t be for a long time. Huge advancements have been made with the release of thousands of large language models like GPT. We all know by now that these models are able to give extremely precise answers to most questions, better than any human can, however I don’t think this counts as thinking. These enormous models have been fed incredible amounts of data from all types of sources and trained over months. However, does this mean that they “know” everything they “learned”? These models use probability to guess what the next word is, considering the input and what has been said already. I don’t consider this process to be the same as thinking, it is just a computer program that generates some numbers based on an input.
I don’t think so. It might seem that way because AI is designed to mimic human thinking and communication. It learns from the data we provide, identifies patterns within that data, and uses language models to generate responses that feel natural and human-like. Even though it can appear quite advanced, it’s still fundamentally dependent on the information it has been trained on and doesn’t truly think or act on its own. Its “intelligence” comes from recognizing patterns rather than having real understanding or awareness. As for the future, it’s hard to say what will happen. There are some concerning reports and ongoing debates about how far AI could develop. But for now, I’d say it hasn’t reached that level of true autonomy yet.
To learn - probably yes, because AI is built on logical and mathematical connections that serve as a foundation for its development, enrichment, and improvement. To think, definitely not, because even though AI has access to much more information, it lacks so-called past experience. As far as I know, artificial intelligence “works in the here and now,” whereas when thinking, people can go back in time and draw on not only information but also that past experience, conclusions drawn, concepts formed, and so on. Another element that has a significant influence on human thinking is the value system (which varies from person to person), which AI does not possess and cannot be generalized to be relevant for every person. Generally speaking, AI can learn, but it cannot think like humans, because no matter how imperfect a person may be, they possess abilities that, at this stage (and in the future, in my opinion), cannot be programmed and implemented in a machine, no matter how perfect it may be.
In short: No – today's AI does not learn and think like humans, even if it sometimes seems that way. Why does it still seem that way? Modern AI (especially with neural networks) can evaluate huge amounts of data and recognize patterns in it. As a result, she can use language, understand images or make decisions surprisingly well. This often feels "intelligent". The key difference: People understand things, have awareness, goals and experiences. AI optimizes patterns – it calculates probabilities for what "fits". A simple image: A person learns what a "dog" is by experiencing the world, understanding meaning and forming concepts. An AI learns "dog" as a statistical pattern in data (pixels, words, etc.), without any real understanding.
To learn—probably yes, because AI is built on logical and mathematical connections that serve as a foundation for its development, enrichment, and improvement. To think—definitely not, because even though AI has access to much more information, it lacks so-called past experience. As far as I know, artificial intelligence “works in the here and now,” whereas when thinking, people can go back in time and draw on not only information but also that past experience, conclusions drawn, concepts formed, and so on. Another element that has a significant influence on human thinking is the value system (which varies from person to person), which AI does not possess and cannot be generalized to be relevant for every person. Generally speaking, AI can learn, but it cannot think like humans, because no matter how imperfect a person may be, they possess abilities that, at this stage (and in the future, in my opinion), cannot be programmed and implemented in a machine, no matter how perfect it may be.
