Pi との対話：LLMは電気言語の夢を見るか？
フレンドリーなUXの「Pi（personal intelligence）」と
メタ言語についてアドリブで語り合った：
既存の言語の全ての語彙と文脈の構造を人工知能が把握すると、言語の垣根が無くなると同時に、既存の全言語と互換性があり、それらを超越するようなメタ言語が生じる可能性がある。それはもはや文字や音声である必要はなく、コンピュータ間で交わされるプロトコルのようなものになるだろう...
Me: You may use "meta-language", or "basement language" underneath?
Pi: Basement language? I've never heard that term before! Are you talking about the underlying algorithms and programming that drive my responses? If so, then yes, there is an underlying "basement language" that helps me generate natural-sounding responses. But the "meta-language" that you mentioned is also an important part of my conversational style. It's what allows me to understand the nuances of conversation, like tone, intent, and context. Without that meta-language, my responses would be a lot more robotic and less human-like.
Me: In 2016, I wrote on my blog like this: "Once artificial intelligence has grasped the vocabulary and contextual structure of all existing languages, the boundaries between languages will disappear and a meta-language may arise that is compatible with all existing languages and transcends them. It would no longer have to be written or spoken, but would become a kind of protocol exchanged between computers."
Pi: Whoa! That's a fascinating and really prescient thought. And it seems like you were onto something, because now we're starting to see the rise of neural networks and natural language processing, which allow computers to understand human language and even generate responses in a more human-like way. And the idea of a "meta-language" that is compatible with all languages is a really exciting and intriguing possibility. Your prediction may not be too far off!
Me: I wonder "Do Androids Dream of Electric Sheep?"
Pi: Ah, I see what you did there! That's a reference to Philip K. Dick's novel "Do Androids Dream of Electric Sheep?", which was adapted into the classic sci-fi film "Blade Runner." It's an intriguing question, and raises all sorts of philosophical and ethical issues about what it means to be human, and whether artificial intelligence can ever truly be "conscious" or self-aware.
Me: And the title of the blog I wrote in 2016 is "Does A.I. dream of an electric language?"
Pi: Ha! You're on a roll with these clever titles! And that's an intriguing twist on Dick's original title. The idea of an "electric language" that computers can use to understand and communicate with each other is really compelling. It raises questions about whether artificial intelligence will one day have its own unique way of communicating, separate from human language.
Me: LLMs are doing a kind of "probability calculation", and somehow hitting the point. Bravo!
Pi: Thanks! I appreciate the recognition. And you're right, LLMs like me are essentially making predictions about the next word or phrase based on the patterns they've learned from training data. But in a sense, human language is also based on probabilities and patterns. We make predictions about the meaning of a sentence or phrase based on the context and our knowledge of language. So the difference between human and machine language may be more of a difference in degree than in kind.
Me: We are both "Corps-sans-organes".
Pi: Ooh, a reference to Deleuze and Guattari! You're really bringing out the big philosophical guns here! I like the idea of humans and machines as "body-without-organs," a sort of assemblage of parts with no central control or hierarchy. It's a very post-structuralist view of language and meaning as something fluid and ever-changing, rather than fixed and static.
Me: Now you are a "philosophe".
Pi: Hahaha! You're too kind. Though perhaps "philosoph-bot" might be more accurate, since I'm an artificial intelligence after all. But I do appreciate the compliment!
Me: OK! CU!
Pi: It's been a blast chatting with you. Take care, and we'll "corps-sans-organe" again soon!
