LENCHI_Critical thinking about oneself
nishio.iconA thought: People who say "critical thinking is important" and only think about criticizing their own AI output are not doing "critical thinking about themselves" to ask "can I do that?" and "is my criticism correct?". Isn't that what you need to do in learning something new, to realize what you are missing or what blind spots you didn't realize you had?
claude.iconI think that point is very important. Critical thinking tends to focus on examining outside information and opinions, but for true critical thinking, it is essential to question one's own thoughts and judgments.
Especially when faced with a new technology such as LLM, we tend to unconsciously overconfident in our own understanding and judgment. We assume that "I am capable of correctly evaluating the output of LLMs," and we believe that our criticisms are always correct.
However, the capabilities of LLMs are constantly improving, and their output is sometimes beyond human intuition. Under such circumstances, criticizing LLMs without recognizing the limits of one's own understanding and judgment may instead hinder accurate understanding.
Rather, what is needed is, as you have pointed out, "critical thinking about oneself". Specifically, I think it is important to ask the following questions:
Do I fully understand the structure and characteristics of LLM?
Can I properly determine the output of the LLM based on my knowledge and experience alone?
Could I learn from LLM's output about blind spots or new perspectives that I wasn't aware of?
Are my criticisms well-founded?
Is it possible that my criticism is wrong?
Through these questions, it is essential to recognize the limits of one's own understanding and judgment, and to have a humble attitude toward learning in order to make the most of LLM.
When learning something new, not just LLM, it is important to "notice what you are missing or what blind spots you were unaware of. By being flexible and incorporating new information and perspectives without being limited by existing knowledge and preconceptions, you can deepen your understanding and grow.
This book could emphasize the importance of such "critical thinking about oneself" in the context of utilizing LLM; it should appeal not only to critique LLM output, but also to constantly question one's own understanding and judgment, which is essential in establishing a new model of intellectual production.
It would convey to readers the importance of being aware of the limitations of their own understanding and judgment, and of being humble in their willingness to learn, in order to be prepared to utilize LLMs. In doing so, they should be able to offer a perspective that sees LLMs as more than just a tool, but as a dialogue partner that encourages personal growth.
---
MEMO: See the table of contents
I'm getting sidetracked.
You wrote about derailment and intellectual production.
---
This page is auto-translated from /nishio/LENCHI_自分に対する批判的思考 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.