Let’s discuss AI in Epistemology. Earth has music for those who listen, and charm and the melody of the earth depend upon the contemporary times’ development. The invention of the computer has been revamping the world in a very new way and where artificial intelligence (AI) is bringing revolution for development. Wait a minute! Is AI real or not? Wow, this is a new pandora of boxes that drag us in a situation where deducing the conclusion from AI myth/fact is a bit harder than can be imagined.
AI has produced many questions, and still, those questions are not answered purely right form at this age of time like can machine think like a human?, will AI defeat human cognition capabilities? and does a machine learn from machine learning and deep learning? etc. Epistemology helps to explore the answer to these questions. Lets first briefly explain the epistemology and role/interest of AI in epistemology for better intuitions.
Epistemology is a branch of science that deals with the theory of knowledge, human cognition, etc. It has helped a lot in the development of AI throughout history because it enables the man to think deep or beneath the surface, explore the answers in a new dimension, and it puts the newly discovered answers for open to discussion. In other words, epistemology gives the theory of AI knowledge in a domain of knowledge validity, scope, justified belief or opinions, etc. It is important in the sense that it delimits and defines the limitations of our cognition system for the development of AI.
3. AI Interest in Epistemology
interest in epistemology and some of that are the following:
- Rationale decisions/thoughts/opinion of human helps the development of AI in a more refined form. Both fields of study are considered as complementary disciplines.
- AI views human instincts or innate behavior from the perspective of formal and statistical computation paradigm that builds the AI relation with epistemology.
- Epistemology opens the gates of knowledge for AI and induces a dynamic approach for AI rather being static, and dynamic approach is supported by machine learning, deep learning, Data Science, etc.
Epistemology is generally divided into four branches and each branch has different views of thought constructions and each poses different kinds of challenges.
According to this concept, things have absolute existence independent of thoughts, ideas, and consciousness, etc. So this absolute existence can be tapped because AI forces man to think computationally and fetch some results from it. One of the biggest challenges that realism poses is how to interpret real existence into the subconscious of the machine. So in this way, the machine can’t be truly intelligent in full potential because there is no instrument or mechanism available yet that translates independent existence into machine consciousness.
According to this concept, a person knows things that exits in minds only. The question arise in mind that how to fetch knowledge from mind because knowledge only resides in minds. Human may know the sky but can’t explain every aspect of it. So the branch of epistemology, idealism, poses a challenge in this way.
This field contends that most of the knowledge can be viewed in terms of practical usage. Only practical ideas are only accepted and non-practical ideas are rejected according to this approach. What about the non-practical things that norms, values, sentiments, and beliefs, etc. If someone is not practicing religious order, it can’t be said that he has no religion. So this approach poses also a challenge.
This type of school of thought has different paradigm of thinking. Accord- ing to this thought, it depends on the human that ’how he constructs knowledge.’ Each individual has different kinds of knowledge. So how a machine can think like a human if the machine has to take/consider different kinds of construction of thoughts. According to one’s view, a picture can be of tiger and other one say
it’s a cat whereas the third person can oppose both and can say that pic is a fiction. Construction of thoughts in AI poses different kinds of challenges.
Epistemology and AI are both complementary disciplines. AI can’t exist without epistemology because it has given birth to AI and for the development of AI, epistemology is most crucial. Human minds also have the certain limitations that limitation does not let humans think beyond the limit. Whereas AI has some advantages besides challenges and vulnerabilities that it has the capacity to envision the things beyond the human limit because the machine has the capability to calculate millions of data within the time of eye blink and compute results within that time.
You may also know: Interoperability of AI and ML systems