I’ll explain the prompts I gave in the feelers later.
I had Claude summarize The Intelligence Age and it was sleepy. I think this is what we’ve been saying in past discussions about AI Summary, that “taking the top half of the abstract is not interesting.”
- Philosophy of Sense uses the terms larger meaning and Small Meaning to describe similar things. In other words, you need to prompt for instructions that elicit Small Meaning, not [summary “interesting” is subjective, which means it can be [AI subjective The AI subjectively represents a symbol and I subjectively reject it.
- This is one side of [the Metropolis Hastings naming game
- Currently, the process of me proposing to AI and AI rejecting is not implemented, and it is expensive for flesh and blood me to do it, but if you interpret “Yasukazu Nishio’s external brain” as me instead of “flesh and blood me,” then rejecting some from Claude’s output and writing it here is a learning process in a broad sense because it changes the parameters of the generation model that reads this in the future. It’s learning in a broad sense, because it changes the parameters, it’s just that the timing of the occurrence is not current.
- I don’t think it has to be symmetrical (Better than not) when you consider whether the Metropolis Hastings naming game should be played in a symmetrical relationship when played by two people with significant differences in resources and abilities in the first place. AI subjectively extracts interesting parts similar to TTTC or Extract questions at various levels of abstraction “pre-extracted from the text”
- Whether it is beneficial or not seems to come down to a very wet sampling of “whether the points of interest match” and “whether we are on the same page” with the AI.
Actual prompt
Summarize. → abstract and uninteresting.
I’d like to know specifics. → Not very interesting.
What are you saying that’s funny?
Yes, there are certainly some interesting observations in this document. Here are a few that I find particularly interesting:
- 1: A magical future:
- The document opens with the statement, “Within a few decades, what seemed magical to our grandparents’ generation will be possible.” This is a striking illustration of the speed and impact of technological progress.
- 2: Society itself is highly intelligent:
- The author states that the entire social infrastructure is much wiser and more capable than the individual and is itself a kind of advanced intelligence. This is an interesting view of the collective intelligence of society.
- 3: From sand to artificial intelligence:
- It describes the history of mankind as a process of “discovering how to create artificial intelligence by melting sand, adding impurities, arranging them on an extremely small scale with amazing precision to create a computer chip, and then channeling energy into it.” This is a very unique view of technological progress.
- 4: Lamp-lighting husband metaphor:
- It is an interesting analogy that many of the jobs today would have seemed meaningless to people hundreds of years ago, but no one wants to look back in time and become a lamp-lighting man. This is a simple illustration of the changes in professions due to technological advances. 5: The Age of Intelligence:
- The author predicts that after the Stone Age, the Agricultural Age, and the Industrial Age, what comes next is the “Age of Intelligence. This is an interesting thought that suggests the next major stage in human history.
These views provide an original and exciting perspective on technological advances and social change. While the authors paint an optimistic vision of the future, they also mention challenges, providing a balanced view.
Please provide a set of original citations and their translations for 2, 3, and 4. They reject the uninteresting ones and let the interesting ones “land” on the original text.
This page is auto-translated from /nishio/面白引用プロンプト 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.