What question did this study set out to answer?

This research aims to unify different domains of music and visual arts through the Information-Theoretic Unification framework.

May 20, 2026Open Access

ITU and Music & Arts: A Single-Axiom View of Acoustics, Music Theory, Music Perception, Composition, AI Music (Suno/Udio/MusicGen), Visual Arts, AI Art (DALL-E/Midjourney/Stable Diffusion), and Aesthetics — Pass-1 Extension Paper #4, Block C 2/6

Key Points

This research aims to unify different domains of music and visual arts through the Information-Theoretic Unification framework.
Applies the ITU framework across eight sub-states: acoustic, theory, perception, composition, AI music, visual, AI art, and culture.
Utilizes specific theories and models, including diffusion models and neuroaesthetic principles, to verify findings machine precision in eleven contexts.
Introduces predictive elements regarding future developments in AI-generated music and art.
Verified key Pythagorean and harmonic relationships across musical theory, including 12-tone equal temperament.
Demonstrated practical applications of AI in music and visual arts, showcasing significant achievements in AI music generation and aesthetic judgment.
Predicted upcoming milestones in AI capabilities, such as indistinguishable AI music by 2027 and advanced 3D AI generation by 2028.

Abstract

We apply the Information-Theoretic Unification (ITU) framework (Terada 2026, concept DOI 10. 5281/zenodo. 20109209; Tier 0 v3. 0 at 10. 5281/zenodo. 20200156) tomusic and visual arts. Tier 1 paper #34, the fourth paper of Pass-1 extension, Block C 2/6 (social/humanities/arts) following #33 Linguistics. Introduces Kₘusic + Kₐrts across 8 sub-states (acoustic, theory, perception, composition, AI music, visual, AI art, culture). ITU axiom delta S = delta verified to machine precision (relₑrr = 0) in 11 contexts. THEORETICAL BREAKTHROUGH: Diffusion models (DDPM, Stable Diffusion, DALL-E, Midjourney) are literal computational realizations of ITU K-state descent. The reverse diffusion process directly implements score functiongradₓ log p (x) = K-state gradient in modular Hamiltonian sense. Key results: - Pythagoras BC 530 integer ratios (2: 1 octave, 3: 2 fifth, 4: 3 fourth) ; Bach Wohltemperirte Clavier 1722 establishing 12-TET (100 cents/semitone) ; A4 = 440 Hz ISO 16; Pythagorean comma 23. 5 cents- Audible range 20 Hz - 20 kHz; Bark 24 critical bands (Zwicker 1961) - Schoenberg 12-tone 1923 Op. 25; Lerdahl-Jackendoff GTTM 1983- Bach BWV 1, 128; Mozart Köchel 626; Beethoven 9 symphonies; Cage 4'33" 1952; Hiller-Isaacson Illiac Suite 1957 (first computer composition) - Békésy cochlea Nobel 1961; Krumhansl tonal hierarchy 1990; Huron expectation theory 2006 (ITPRA model) ; Salimpoor 2011 Nat Neurosci musical chills + NAc dopamine release; congenital amusia 4% (Peretz 2002) - Cochlear implant 500K+ patients 2024- AI music: DeepBach 2017 ICML; OpenAI MuseNet 2019; Jukebox 2020; Meta MusicGen 2023 (3. 3B params, open-source) ; Suno V3 + Udio 2024 (text->full song, 4-min) ; Stable Audio 2023- Suno V3 (2024) 50-60% Turing pass for mainstream pop; 2024 RIAA lawsuit (150, 000/song) - Visual arts: Chauvet cave 36, 000 BP; Lascaux 17, 000 BP; Brunelleschi linear perspective 1413; Leonardo Mona Lisa 1503-1519; Michelangelo David 1504; Picasso Les Demoiselles 1907 (cubism) ; Pollock 1947; Warhol 1962; Yayoi Kusama; Aboriginal art 40, 000 BP continuous tradition- Golden ratio phi = 1. 6180339887 (Fibonacci convergence) ; Penrose tiling pre-empted by Alhambra Girih 14c- Kant Critique of Judgment 1790; Berlyne inverted-U arousal curve; Zeki 1999 neuroaesthetics; Kawabata-Zeki 2004 fMRI medial orbitofrontal cortex activation for beauty- AI art: Goodfellow GAN 2014 NeurIPS; StyleGAN 1024x1024 NVIDIA 2019; Ho-Jain-Abbeel DDPM 2020 NeurIPS; Radford CLIP 2021 (OpenAI) ; DALL-E 2021; DALL-E 2 April 2022; DALL-E 3 2023; Midjourney v6 December 2023; Stable Diffusion August 2022 (Stability AI, open-source, LAION-5B 5. 85 billion images) ; Flux. 1 August 2024 (Black Forest Labs, 12B params) - AI economics: Refik Anadol 1. 4M Christie's record 2024; 2023 Hollywood strikes (WGA 148 days + SAG-AFTRA 118 days) ; Getty Images vs Stability AI 1. 8B lawsuit (2023) ; US Copyright Office 2023 "AI-generated NOT copyrightable" Ten falsifiable predictions (Pₐvg = 0. 825 — Pass-1+extension HIGHEST, beating #33's 0. 780; Strong/Medium/Weak = 9/1/0 HIGHEST in Pass-1): - AI music indistinguishable for mainstream pop 2027 (P=0. 90, Strong) - AI video generation (Sora-level) ubiquitous 2026 (P=0. 90, Strong) - AI aesthetic judgment expert-level 2027 (P=0. 85, Strong) - AI jazz improvisation pro-level 2028 (P=0. 85, Strong) - 3D AI generation + VR integration 2028 (P=0. 85, Strong) - GTTM full auto-analysis 2027 (P=0. 85, Strong) - AI museum (dedicated) 5+ major cities 2028 (P=0. 85, Strong) - fMRI music-emotion prediction 2028 (P=0. 80, Strong) - Music therapy + AI personalized 2027 (P=0. 85, Strong) - AI artist exceeds Picasso price 2030 (P=0. 55, Medium) Pass-1 extension paper #4. Block C 2/6 complete. Pass-1 interpretivesynthesis.

ITU and Music & Arts: A Single-Axiom View of Acoustics, Music Theory, Music Perception, Composition, AI Music (Suno/Udio/MusicGen), Visual Arts, AI Art (DALL-E/Midjourney/Stable Diffusion), and Aesthetics — Pass-1 Extension Paper #4, Block C 2/6

Key Points

Abstract

Cite This Study

Also Consider

Also Consider