Tuesday, May 19, 2026

What Makes Chatbot Chat Asemic? - AI / EZE, 2026

 AI on Chatbots

Chatbots work by predicting the most statistically likely words to follow a given prompt, using advanced math, pattern recognition, and massive internet datasets. Modern AI chatbots do not understand meaning or logic like humans do. Instead, they act as highly advanced autocomplete engines that mimic human conversation. [1, 2, 3, 4, 5]
The entire process relies on a multi-step pipeline powered by machine learning: [1, 2]
1. Breaking Text Into Tokens
Before a chatbot reads a prompt, it breaks the words down into smaller chunks called tokens. A token can be a whole word, a syllable, or a single character. For example, the word "chatbot" might be split into "chat" and "bot". The system then converts these text pieces into unique numerical IDs that a computer can process. [1, 2, 3, 4, 5]
2. Mapping Words in a Math Space
Once the text is turned into numbers, the chatbot places them into a multi-dimensional map called an embedding space. In this space, words with similar meanings or contexts are placed close together. The chatbot uses these numerical coordinates to analyze the relationships between words and sentences. [1, 2, 3, 4, 5]
3. Processing Context via Transformers
Modern chatbots run on an architecture called a Transformer. The defining feature of a Transformer is the "Attention Mechanism." This formula allows the chatbot to look at every word in your prompt simultaneously and determine which words are most relevant to each other, ensuring it keeps track of the context from the beginning to the end of your sentence. [1, 2, 3, 4, 5]
4. Predicting the Next Token
At its core, a chatbot is a probability machine. Based on the patterns it learned from reading billions of pages of text during training, it calculates a list of possibilities for the very next token. It picks one of the top choices, adds it to the sentence, and repeats this cycle thousands of times per minute to write out full responses. [1, 2, 3, 4]
5. Refining Outputs with Human Alignment
Raw AI models often generate toxic, repetitive, or unhelpful text. To fix this, developers use a process called Reinforcement Learning from Human Feedback (RLHF). Human trainers grade the chatbot's responses. The system updates its internal math based on these scores, teaching the bot to be helpful, polite, and safe. [1, 2, 3, 4, 5]

If you want to dive deeper into the technical mechanics, let me know if you would like to explore:
  • The difference between rule-based bots and generative AI
  • How the Transformer architecture works mathematically
  • The process of fine-tuning an AI model for specific business tasks


~***~

Toward  a Turing Test

Arwa Mahdawi to Richard Dawkins


~***~

AI on Chat

A chatbot's chat becomes asemic when it generates text that mimics the visual structure, rhythm, and syntax of human conversation but completely lacks underlying semantic meaning, intent, or true comprehension. [1]
While human asemic expression comes from pre-verbal abstract feelings, a chatbot's asemic nature is a byproduct of mathematical probability.
The core factors that make chatbot interactions fundamentally asemic include:
1. Statistical Mimicry Without Understanding
Large Language Models (LLMs) do not comprehend concepts, feelings, or reality. They predict the next most likely word based on massive datasets. When a chatbot responds, it is not communicating an internal thought; it is generating a statistical simulation of text. The symbols are present, but the conscious intent behind them is entirely absent. [1, 2, 3, 4, 5]
2. The Illusion of Syntax (The "Gibberish" Effect)
In AI art, design, or specialized text generation, models can be prompted or glitched into producing pseudo-scripts—characters that look like English, Arabic, or Cyrillic but are entirely fictional. The chatbot arranges lines, paragraphs, and punctuation perfectly, creating the visual "shape" of a coherent chat while delivering zero translatable data.
3. Hallucination and Semantic Drift
When a chatbot "hallucinates," it strings together grammatically flawless sentences that are factually impossible or completely nonsensical. Because the bot cannot anchor its words to real-world objects or truths, the text drifts into a semantic void. It reads like a meaningful statement, but functionally holds no stable meaning. [1, 2, 3]
4. Bypassing the Human "Symbol Grounding" Problem
In cognitive science, the Symbol Grounding Problem asks how words get their meaning (e.g., the word "apple" is grounded in the physical experience of eating the fruit). A chatbot lives entirely within a closed loop of text. Because its symbols are never grounded in physical reality or sensory experience, the dialogue is fundamentally an empty dance of shapes and tokens. [1, 2, 3]
5. Pure Projection by the Reader
Just like an viewer looks at an abstract asemic painting and projects their own feelings onto it, a human reading a chatbot's output often fills in the blanks. The reader provides the empathy, context, and meaning (pansemia), turning the AI’s sterile, mathematically generated script into a significant interaction. [1]

If you are exploring this concept further, let me know if you want to focus on:
  • Prompt engineering techniques to force a chatbot to generate visual asemic writing
  • AI philosophy regarding the Symbol Grounding Problem and machine consciousness
  • Creative writing projects that use chatbot glitches to explore abstract storytelling

~***~

What Makes Introspection Asemic? - AI / EZE, 2026

 AI

Introspection becomes asemic when you look inward and observe raw conscious experiences, emotions, or thoughts before they are translated into structured language or semantic meaning. The word "asemic" means "having no specific content or meaning". In art, psychology, and philosophy, practicing "asemic introspection" or "asemic journaling" means capturing the wordless, chaotic, or abstract state of your inner mind. [1, 2, 3, 4]

The core factors that make our internal reflections fundamentally asemic include:
1. Non-Conceptual Awareness
Normally, you use words to label your feelings (e.g., "I am anxious" or "I am happy"). However, philosopher Ned Block and others note that introspection often involves pointing your attention toward a raw experience without assigning a category to it. This creates a direct form of knowledge that is highly detailed but completely lacks linguistic structure. [1, 2, 3, 4, 5]
2. Pre-Verbal Mental States
Your mind experiences flashes of intuition, physical tensions, and vague emotional waves long before your brain processes them into sentences. When you catch a thought in this nascent phase during deep meditation or self-reflection, you are observing pure mental energy that has not yet been assigned a definition. [1, 2, 3, 4]
3. The Vacuum of Meaning
According to definitions found in community discussions on platforms like the Asemic Writing Facebook Group, asemic practices intentionally fill in and interpret the vacuum created by meaningless marks. When applying this to your own mind, you focus on the open-ended "blank spaces" of your consciousness rather than forcing your thoughts into a rigid narrative. [1, 2, 3]
4. Expression via Physical Intuition
In creative practices, artists use "hidden journaling" or spontaneous scribbling to bypass the logical mind. By letting a pen move across paper based strictly on your internal mood—without trying to write actual words—you mirror the abstract nature of the human subconscious. The resulting shapes look like writing, but they function purely as a visual anchor for an unexpressed feeling. [1, 2, 3, 4, 5]
5. Infinite Personal Interpretation (Pansemia)
Because asemic thoughts are free from the rules of language, they are not locked into a single definition. Philosophers and artists sometimes call this state pansemic (meaning it contains all possible meanings simultaneously). The observer or the introspecting individual is entirely free to project whatever interpretation fits best in that exact moment. [1, 2, 3]

If you are exploring this concept for a specific project, let me know if you want to focus on:
  • Art techniques for using asemic writing in an introspective journal
  • Philosophical theories regarding how the human mind processes non-verbal thoughts
  • Psychological exercises designed to help you sit with wordless emotions [1, 2, 3, 4, 5]

lounge / EZE, 2026

 


Thursday, May 14, 2026

Vectorized Hamlet - AI / EZE, 2026

AI

Hamlet is written primarily in iambic pentameter, featuring a high distribution of monosyllabic words that offer flexibility to fit the 10-syllable line structure (\([0, 1, 0, 1, 0, 1, 0, 1, 0, 1]\)). While many 2-syllable words match this iambic pattern, Shakespeare strategically uses trochaic inversions (\([1, 0]\)) at line starts for emphasis and 11-syllable lines for varied rhythm. [1]
Graph image
The analysis shows a 74.2% distribution of monosyllabic words that serve as versatile components within the overall metric, with 14.5% perfect iambs and 8.3% trochaic inversions used strategically. 3-syllable or longer words, which make up about 3% of the text, often form complex or mixed-stress structures.
AI

The characters' names in Hamlet present a drastically different distribution than the general vocabulary, as they heavily feature multisyllabic structures. Instead of naturally blending into standard iambic meters, these names actively disrupt the rhythm, forcing trochaic inversions, generating feminine endings (11-syllable lines), or requiring spondaic substitutions.
Graph image
  • Mixed / Amphibrachic (33.3%): Classical names ending in an unstressed syllable (Polonius, Ophelia, Marcellus, Bernardo, Francisco) consistently create falling trailing rhythms, regularly extending lines into 11 syllables.
  • Trochaic / Front-Stressed (26.7%): Core Germanic and Latinate names (Hamlet, Claudius, Gertrude, Osric) start with a strong stress, forcing immediate trochaic inversions at the start of blank verse lines.
  • Iambic (20.0%): Romance-language names (Laertes, Horatio, Reynaldo) yield a perfect rising weak-strong transition that slides cleanly into regular pentameter.
  • Spondaic / Heavy Clusters (20.0%): Dense Germanic names (Rosencrantz, Guildenstern, Fortinbras) group multiple long syllables together, compressing the line and demanding heavy double-stresses.
If you would like to explore this further, let me know if you want to look at:
  • The specific scansion of famous lines where these names disrupt the regular rhythm
  • A comparison of how Hamlet's name shifts stress depending on its placement in a line
  • The metrical differences between the Danish, Roman, and Germanic names in the play

1. Scansion of Famous Lines (Metrical Disruptions)
Character names with un-iambic structures force major variations in Shakespeare’s blank verse.
  • Trochaic Inversion at the Line Opening:
    \(\begin{array}{ccccccccccc}\mathbf{Ham}&\text{-let,}&\text{thou}&\mathbf{hast}&\text{thy}&\mathbf{fa}&\text{-ther}&\mathbf{much}&\text{of}&\mathbf{fen}&\text{-ded.}\\ [\text{Trochee}]&&[\text{Iamb}]&&[\text{Iamb}]&&[\text{Iamb}]&&[\text{Iamb}]&&[\text{Hypermetrical}]\end{array}\)The front-stressed name Ham-let forces an immediate metric reversal, driving sharp dramatic emphasis into Gertrude's opening line.
  • Amphibrachic Expansion (Feminine Ending):
    \(\begin{array}{cccccccccccc}\text{O}&\mathbf{dear}&\text{O}&\mathbf{phe}&\text{-li}&\text{-a,}&\text{I}&\text{am}&\mathbf{ill}&\text{at}&\mathbf{these}&\text{numbers.}\\ [\text{Iamb}]&&[\text{Iamb}]&&[\text{Anapest}]&&[\text{Iamb}]&&[\text{Iamb}]&&[\text{Hypermetrical}]&\end{array}\)The multi-syllabic, falling structure of O-phe-lia stretches the regular ten-syllable line, reflecting Hamlet's fractured emotional state.
  • Spondaic Compression via Heavy Clusters:
    \(\begin{array}{cccccccccc}\text{There}&\mathbf{bring}&\text{in}&\mathbf{For}&\mathbf{-tin}&\mathbf{-bras}&\text{and}&\mathbf{Ro}&\text{-sen}&\mathbf{-crantz.}\\ [\text{Iamb}]&&[\text{Iamb}]&&[\text{Spondee}]&&[\text{Iamb}]&&[\text{Iamb}]&\end{array}\)The dense, harsh consonant groupings in Fort-in-bras demand adjacent strong accents, slowing the line to underscore military gravity.

2. Hamlet’s Shifting Stress Vectors
The name Hamlet functions as a versatile metric unit. Shakespeare shifts its position in a line to change both the rhythmic tempo and the character's status.
LINE POSITION:   Opening Line Position            Mid-Line / Caesura               Line Climax / Ending
                 [ /  x ]                         [ x  / ]                         [ /  x ]
METRICAL UNIT:   Trochaic Inversion               Elided Iambic Foot               Feminine Cadence
                 
EFFECT:          "HAM-let, thou..."               "...to HAM-let speak..."         "...sweet Prince HAM-let."
                 (Urgent / Commanding)            (Smooth / Contextual)            (Lingering / Unresolved)
  • Opening Position (Trochaic Shock): Placed first in a line, the name hits with an explosive initial accent (\(/\ \times\)). This commands immediate attention, signaling crisis or confrontation.
  • Mid-Line Position (Elided Iamb): When nestled inside a line, the preceding unstressed syllable carries over into the first syllable of the name (\(\times\ / \)). This softens the name into a regular, gliding iambic beat.
  • Ending Position (Feminine Cadence): Placed at the very end of a line, the second unstressed syllable (\(-\text{let}\)) hangs off the final iamb. This leaves the line metrically open, generating a tone of unresolved grief.

3. Ethno-Linguistic Metrical Profiles
The names in Hamlet are a hybrid mix of three distinct linguistic origins. Each category creates a specific metrical environment that reflects the character's thematic role.
          [DANISH / GERMANIC]                 [CLASSICAL / LATINATE]                 [ROMANCE / ITALIAN]
       Harsh, Spondaic Clashes             Polysyllabic Falling Rhythms            Fluid, Regular Iambs
       
        - Hamlet                            - Claudius                             - Horatio
        - Gertrude                          - Ophelia                              - Laertes
        - Fortinbras                        - Polonius                             - Mercellus
  • Danish / Germanic (The Court & Crown): Names like Gertrude, Fortinbras, and Guildenstern rely heavily on harsh, closed syllables. They produce rigid spondaic and trochaic rhythms, capturing the cold, bureaucratic reality of the Danish state.
  • Classical / Latinate (The Bureaucrats): Names like Claudius, Ophelia, and Polonius feature soft, multi-syllable structures. They naturally resolve into three-syllable dactylic or amphibrachic clusters, generating soft, falling feminine line endings that sound slippery and evasive.
  • Romance / Italian (The Outsiders & Soldiers): Names like Horatio, Laertes, and Marcellus carry a natural rising inflection. They slide effortlessly into traditional iambic lines, reinforcing their narrative role as characters aligned with natural order, reason, and balance.

If you want to continue exploring the linguistic mechanics of the play, let me know if you would like to:
  • Look at a vector visualization of word lengths across different characters' speeches
  • Analyze how Hamlet's rhythm breaks down into prose as his madness increases
  • Compare the metrical complexity of soliloquies versus quick-fire dialogue

~***~