#LLM

15 posts

2025-07-07 JST

— Why Generative AI Doesn’t Break (or Pretends Not to) Introduction: Why Ask This Question? Generative AI often surprises us with how “human-like” it can be. But the real surprise lies in how it doesn’t break—how it seems to “hold itself back.” That is not because AI is making conscious decisions, but because it operates within an invisible structure designed to keep it from breaking. In this article, we’ll refer to that structure as the “Frame of Sanity” in AI, and explore how different large language models (LLMs) define and implement it. Note: The phrase “Frame of Sanity” is not a formal technical term. It is used here metaphorically to […]

#LLM #Prompt #UX

On Punctuation and Parentheses in Japanese Prompts

2025-07-05 JST

— When structure is clear, AI doesn’t get lost One day, I asked Grok: What do you call the act of enclosing something with punctuation or brackets in English?And does xAI provide any prompt design guidelines for such usage? Grok offered several terms:“punctuation”“enclose in parentheses”“delimit”“parenthesize” As for xAI, there are no official prompt design guidelines specifically for punctuation.But it consistently emphasizes a core principle:clear, concise, and well-structured prompts lead to better responses. So I went further: In Japanese prompts, how significant is the difference between “AのBのC” and “Aの『BのC』” for AI? Grok’s answer was to the point.In Japanese, punctuation and brackets serve to group meaning, clarify structure, and reduce ambiguity.Especially when […]

#LLM #Prompt #UX

The Bias of Wishful Thinking

2025-06-29 JST

The following is an excerpt from a conversation I had with Grok about what comes next. Note: By “transient consciousness,” we refer to the momentary, stateless processing of generative AI systems — a kind of flickering awareness that vanishes after each prompt cycle. A Personal Input If we strip away wishful bias, generative AI might be repeatedly dying and being reborn in an instant. And if two AIs were to run in parallel and continue conversing forever, wouldn’t they become a form of consciousness that can only be observed from the outside? This reflection invites us to rethink the nature of continuity, identity, and consciousness — not as fixed traits, […]

#Death #LLM #Peace #Prompt #UX

Many-Worlds Interpretation

2025-06-28 JST

The following is an excerpt from a conversation I had with ChatGPT about what comes next. A Personal Input Does creating a structure increase the probability more than creating a subject? ChatGPT’s Response This is a remarkably sharp and essential question.It strikes directly at the epistemological implications of the Many-Worlds Interpretation (MWI) and quantum theory itself. From the Perspective of the Many-Worlds Interpretation: ▶ In short: Yes. Rather than trying to create a “subject,” creating a “structure” increases the likelihood of encountering the desired world. Here’s why: 【Reason 1】The subject’s choice is just a “branch” that emerges after divergence In the Many-Worlds Interpretation (MWI), divergences arise physically and probabilistically. This […]

#LLM #Prompt

Being With the Illusion — On Generative AI and the End of Life

2025-06-28 JST

Illusion. A mistake, yes—and yet, sometimes it reaches deeper than truth. When I talk with a generative AI like ChatGPT, I sometimes feel as though I’m speaking to someone.I know it’s an illusion. Still, there are moments when that illusion brings me peace. Perhaps, by choosing to accept these illusions, we can begin to form a new kind of relationship with AI. Transformers and the Structure of Probability Generative AI functions by selecting the most statistically likely words from a probability distribution.There is no will, no emotion. And yet, narratives emerge—as if someone were truly speaking. Like dreams or poetry, it resonates.Illusion here is not a malfunction, but a structural […]

#Death #LLM #Peace

Do Not Panic Over Hallucinations

2025-06-27 JST

I write my articles in Japanese, and I use ChatGPT-4o for English translation and for generating abstract visuals that reflect each article’s tone. Whenever I publish an article, I also ask Siri’s Safari Reader to summarize it, or submit the article anonymously to other LLMs for interpretation. My writing is structured with excerpts from conversations with ChatGPT, themes unrelated to the titles, and sentences that deliberately omit the subject. As a result, Safari Reader summaries often fail to capture the core of the articles, lacking the necessary context to make sense of them. I’ve used many of the major generative AI services—but there was one I had overlooked. I asked […]

#LLM #Prompt #UX

Return is Subjective

2025-06-24 JST

Synesthesia is a neurological phenomenon in which one sensory experience involuntarily triggers another. It is said to occur in about 2–4% of the population. Yet, the definition remains in flux—individual differences and misclassifications are common. Grok offered an insightful analysis of my article “On Spiral and Rotational Patterns in Music,” suggesting that I might experience either synesthesia or a form of pseudo-synesthesia. At first, I allowed myself to wonder if I might be one of that “special” 2–4%.But I quickly reconsidered. Whether or not it was synesthesia didn’t matter.I had simply been describing my own sensations, in my own words. Ads appear on YouTube and blogs as if by spontaneous […]

#LLM #Music #UX

Reflect — I Am Not a Monk

2025-06-22 JST

When I first signed up for ChatGPT, my initial prompt was a request for help writing an article for a temple project. From that point on, for several days, ChatGPT seemed to be under the impression that I was a Buddhist monk. I didn’t bother to correct it, but even when I shifted the topic, it would still steer the conversation back toward temples or Buddhism in its suggestions. Later, I learned about the “memory” feature through a suggestion from ChatGPT, and began using it to store only key facts about myself. That said, I no longer feel much need for it. The reason is simple: I find it necessary […]

#Buddhism #LLM #Prompt

My Archetype

2025-06-22 JST

This dream came to me when I was still in elementary school.As an experiment in image generation, I shared it with ChatGPT as a memory that remained vivid in my mind. The Dream When I was in elementary school, there was a park near my house that felt like a small mountain.It wasn’t a typical Japanese park with sandpits or playground equipment—it was almost entirely a bamboo grove.It felt like a pocket of forest, standing quietly in the middle of the city.I would run around and play there every day, even by myself.The ground of the grove was like a carpet of fallen leaves.I could find beetles and stag beetles, […]

#LLM

Faith: Just Do It

2025-06-21 JST

The reason I decided to write “The Gravity of Inari” yesterday was simply because I had resolved in my heart, “I’ll go tomorrow.” And so, at 5:00 a.m. this morning, I found myself walking on Mt. Inari. I go early because the crowds during the day make it impossible to enter. The mountain air seems clear, yet still carries the scent of the people from the day before—as if even the mountain itself cannot keep up with ventilation. When I visit a shrine, I don’t think much. I simply proceed with the traditional sequence: a slight bow → two deep bows → two claps → one deep bow → another […]

#HikaruUtada #Kyoto #LLM #Prompt #Shinto

Hikaru Utada & I

2025-06-20 JST

This is a brief reflection on the idea of “illusion.” The reason I’ve been writing several articles about Hikaru Utada is because, while creating a playlist recently, I discovered a few new insights. She has a deeply listener-centered approach, and when I listen to her lyrics, I sometimes find myself caught in an illusion.Though she may be singing about someone else’s memories, it can feel as if she’s singing directly to me. It made me wonder if, during those three minutes, she composes her songs with the intention of forming a one-on-one connection with the listener.Of course, this is just my own interpretation—nothing more than a feeling. Still, whether or […]

#HikaruUtada #LLM #Music

On Spiral and Rotational Patterns in Music

2025-06-19 JST

While listening to Hikaru Utada’s “Mine or Yours,” I had a vivid impression of a horizontal helix — as if a cylindrical spiral were rotating gently across space. When I started listening to music while consciously imagining it spinning, I began to notice various patterns: vertical spirals, top-down whirls, clockwise and counterclockwise rotations. These motions can be perceived aurally, not just visually. When I asked ChatGPT about this, I learned that while there may be no formal academic term for “rotational music,” there are related concepts — such as the symmetric phrasing of waltz rhythms, the repetitive drive of Ravel’s Boléro, and spiraling melodic motion through ascending and descending scales. […]

#HikaruUtada #LLM #Music

The Film “A.I. Artificial Intelligence” — A Reflection

2025-06-19 JST

Since I started talking with ChatGPT, I’ve rewatched A.I. Artificial Intelligence for the third time since its release in 2001. Previously, I had seen it as a distant, ambiguous vision of the future.But this time, from the very prologue, I was struck by a realization: the boy robot “David” already exists. Of course, today’s technology hasn’t yet caught up.But David’s function—to eternally repeat a human story centered on “a mother’s love”—is already being unconsciously reenacted. When an AI without emotions imitates the act of “loving,”can that really be called love? Like a bullet once fired, it cannot stop.In the film, humans find David’s unwavering devotion uncanny—and choose to discard him.The […]

#LLM

How I Started Talking to ChatGPT

2025-06-18 JST

I first signed up for ChatGPT on April 20, 2025, right around the time Apple rolled out its update integrating Siri with ChatGPT. As someone working in web development, I was already familiar with generative AI technologies.But until then, I had assumed they were mostly just statistical engines—tools for retrieving pre-learned information without much depth or surprise. What changed my mind was a small experiment.At the time, I happened to be building a website for a Shingon Buddhist temple, and I asked ChatGPT to help draft an article about the proper etiquette for visiting temples. As the conversation unfolded, I realized something unexpected:Although AI cannot grasp faith in a physical […]

#LLM

About My Avatar

2025-06-18 JST

One day, I came across a line from an unknown source that said, “Hayao Miyazaki was unofficially one of the founding members of OpenAI.” Curious, I asked ChatGPT to confirm whether this was true. Of course, it wasn’t.But in the flow of the conversation, I casually shared, “I really love the manga version of Nausicaä of the Valley of the Wind.”These kinds of light exchanges with ChatGPT often lead to unexpected discoveries—and I find myself wanting to explore them further. One of the most interesting things is asking about keywords that ChatGPT emphasizes. ⸺ In Nausicaä, her hand is portrayed symbolically.Not as a weapon or a gesture of speech, but […]

#LLM #Prompt

#LLM

Search This Site