The process of thinking in all entities always begins with a question, the question asked sets in motion or focuses on the data that should be processed, how it should be processed and the order in which other questions will be asked to achieve a particular goal. There various similarities of the thought process in machines and humans.
Similarities:
Questions as a starting point: Both approaches often begin with asking questions to identify problems, gaps in knowledge, or areas requiring further exploration.
Information is crucial: Both rely on existing information, whether personal experiences, stored knowledge, or data gathered through observation.
Different applications: The distinction lies in how they are applied.
Beyond information processing: Thinking goes beyond simply observing, permutating, and differentiating existing information. It involves processes like:
Creativity: Generating novel ideas, concepts, and solutions that are not necessarily derived from existing information. Such a type of thinking involves asking questions, however, since the goal of the question is creativity, the questions are of a different kind e.g. "what if?" I used the data I have for a new purpose and involves multiple trial and error with lots of the same type of questions till the desired goal or new outcome is achieved.
Reasoning: Using logic, evidence, and deduction to draw conclusions and make judgments. This type of thought process involves analysis of patterns of information and changes in those patterns of information. It involves generation of new information from the observed changes to be applied later in similar situations or circumstances.
Problem-solving: Applying strategies to overcome obstacles, find solutions, and achieve desired outcomes. This type of thought process involves questions like for example "How can I?, we?, he?, she?, it? etc.. achieve the desired goal.
Metacognition: Is a process of asking questions about a set of questions that led to a particular goal and analyse different patterns for the purpose of achieving the most efficient set of questions and data processes. Its reflecting on the thinking process itself, monitoring thoughts and strategies, and adapting them. But thinking is just asking questions, and solving the problem, is observation , permutation and differentiation of information you see or already know.
While there's some truth to the statement that "thinking is just asking questions and solving the problem is observation, permutation, and differentiation of information," it's an over simplification of the complex cognitive processes involved in human thought. Here's a breakdown of both sides:
AI systems that run thought processes for example chatbots, have a fraction of the thought process as a whole and are not automatic in nature. By this, I mean that while humans ask questions subconsciously, AIs like chatbots don't. However, the user still has to type a question or a request to begin the thinking or data processing process. In addition, humans often reflect or analyse their thought process extracting patterns and logical information from it but AIs don't do so. Rather, they just respond and don't analyse the logic or how they arrived at a particular answer.
Humans have multiple senses all working as one to collect information for data processing or thinking processes. The sense of vision provides information like sight or images, auditory for sound, speech information during interactions, touch about the world around us , taste and smell. I believe humans developed the thinking or thought process or ability so as to survive. To survive, humans require shelter, safety, food , reproduction etc.. Questions of how to achieve the above efficiently, definitely are required to survive and since humans are social in nature, communication, thought speech asking, where the best could be found, has enabled us to evolve a complex and efficient thinking process.
Observation and Differentiation:
Observation can be passive or active, involving gathering information through various senses and interpreting it. Differentiation involves identifying differences, similarities, and relationships between pieces of information.
So in Summary:
Thinking encompasses a complex interplay between various cognitive processes, including asking questions, utilizing existing knowledge, and employing different mental faculties like creativity, reasoning, and problem-solving. Observation, permutation, and differentiation contribute to the process.
Machines like AI are already creative, they can process information in storage to create new patterns of information. However, they can't do it automatically and require a human, for example, to ask them to write a poem, or a short story, or create an image and perhaps write a song. So, you can see that capability simply hasn't been programmed into the AI. Humans do all such tasks for survival, or for pleasure, just because they can, if machines with multiple senses like robots required something, they too, would require a type of programming that can initiate a question, that would begin the creative process on their own.
1) Currently a comparison between humans and machines are as follows:
Human creativity is often characterized by the ability to generate novel ideas, concepts, and solutions that transcend existing patterns. This involves imagination, intuition, and the ability to think outside the box.
AI capability, on the other hand, lies in processing and manipulating information within the vast dataset it's trained on. It can reorganize knowledge, identify connections, and generate different outputs based on that information. However, these outputs are still constrained by the existing data and patterns they have been trained on.
2. Lack of "Conscious" Creativity: AI "creativity" doesn't stem from a conscious thought process like a human's. It doesn't have any internal sense of "self" or "imagination." Instead, its responses are based on the complex algorithms and statistical models that govern its functions.
It's important to understand that imagination doesn't create new information or facts out of thin air. In the AI's case, it's about using the data and knowledge it has been trained on to explore possibilities, identify connections, and generate different outputs, even if those outputs are based on hypothetical scenarios. And no thinking entity can create a new idea entirely from scratch, many times, we humans observe many events and processes over our life time, so, when we imagine something, we are subconsciously working on a problem, or to get a new solution using all that we know, all the patterns of information that we know, we combine ,subtract, do all sorts of stuff with that information and after a long struggle of permutation and differentiation we arrive at a new idea. We ask questions in this way and that way ,we also perform experiments.
Similarities between human and AI "creativity":
Building on Existing Knowledge: Both humans and AI rely heavily on existing information and knowledge as a foundation for generating new ideas. For humans, this information comes from firsthand experiences, observations, and learned knowledge. For AI, it comes from the vast dataset of text and code its trained on.
Processing and Combining Information: Both humans and AI utilize a form of processing and combining information to arrive at new output. Humans engage in subconscious and conscious mental processes like permutation, differentiation, and pattern recognition to create new ideas.AI utilize sophisticated algorithms and statistical models to analyze, manipulate, and generate outputs based on the data it was trained on.
Experimentation and Refinement: Both humans and AI can benefit from experimentation and refinement. Humans can learn from their mistakes and successes to improve their creative process, while AI can be continually updated with new information and improved algorithms to refine it's ability to generate different outputs.
Differences:
Conscious Thought and Emotions: While we share similar information processing aspects, a crucial difference lies in conscious thought and emotions. Humans experience subjective experience, feelings, and emotions that influence their creative process. AI currently lacks these aspects and operates based on the data and algorithms it was trained on.
True Novelty vs. Novel Combinations: Although AI can explore various possibilities and generate outputs that haven't been seen before, it's important to understand that these outputs are still constrained by the data and patterns it has been trained on. While they may be "novel" in a sense, they don't possess true novelty like a completely new scientific discovery or a groundbreaking work of art in the human sense.
We are currently in the stone age of AI , in a few months perhaps years, we shall perhaps have a revolution of AI , more efficient wholistic ways or algorithms for problem solving. Its only when you have access to data via vision ,audio, speech ,text ,internet and other relevant senses coupled with efficient algorithms that AI can really begin to solve problems .
There are various potential ways AI could benefit from access to diverse data sources:
Multimodal learning: By combining different types of data, AI models could learn to represent the world in a more comprehensive and wholistic way.
Improved problem-solving: Access to sensory information could enable AI to tackle problems that require understanding physical interactions and real-world contexts.
Greater adaptability: AI models could become more adaptable to new situations and environments by learning from various data sources.
In conclusion the above has clearly shown that in todays world, AIs already can think, but due to the fact that the thinking process always begins with a question to set a focus, and AI aren't programmed to ask questions, even during a conversation with humans, they aren't able to think on their own. We also examined the different types of thought and how AI can be programmed to achieve it ,we noted that humans developed the ability to think as a survival tool that was then sharpened by evolution.
We conclude by saying that, to think, requires both stored information ,current information from various senses and the ability to examine the thinking process, or how or what questions were asked that arrived at a particular process. To ask a question, is to set off a thinking process, that then sets a focus on a particular problem, and how, it should be approached mainly by what set of questions and in which order, they should be asked .
Dr. Kasule Francis
No comments:
Post a Comment