this is the fifth time today my eyes fall closed.
in a minute or two i'll dissolve into the night sky.
i've become much more silent,
but only a little less eager.
i'm tired during her absences,
and she vividly walks in dreams.

this is the third time this week i think of cigarettes.
i wonder how i manage to survive the suffocation.
i'm living with severe headaches,
and loving my personal sadness.
so in turn i put death between my lips,
and slowly breathe my life away.

this is the first time you affect me this way.
and i dug deep in the earth to find you alive.
i want you as much as i don't.
it's like nicotine and sleep;
more than a little and less than too much,
just enough,
until we are airborne for good.

My nostrils stay cold in the warm weather, cold in the rain, a forever remnant of the days of ******* and truck beds. I inhale and exhale the poison of cigarettes, but the shotgun electricity of the little white lines pierces the folds of memory. As much as I ache to forget, I can still feel the powder laying latent underneath my fingernails.

The days of wanting stress are replaced by wanting to alleviate it. I'd rather not sit and listen to your scratching your tense fingertip-tapping jitters. Silent leg shaking bouncing making my records skip. The dust-covered dumpster-dived needle has stress enough without your additions, subtractions, multiplications. You sneeze white and red, the signs of frustrated futures and presents. The record skips back to one, water stained, nothing changes.

I once played without direction and felt it cheap and unnecessary, like angels that breathe deeply underwater. Grasping for sympathy and votes of confidence. Forging intimate connections without it, needing wanting grabbing feebly into the air. Desperation never even gets a chance. We are strong as equals, love dissipating into the aether, waiting watching wandering wishing waking.

I tried to bend and not break, divide and not conquer. I tripped on the wire, skinned knees, forgetting. The clouds of gray hang low low in the air. I will hold strong to my promises, even in this time of turmoil and smashed faces. My foot will stay planted. I will move forward. I will keep on keep on keep on.

Even when they doubt me, I will keep on.

I'm good at math
Whenever my mind gets bored
i practise multiplications
give myself math problems
I always solve them
correct

I tend to remember numbers
1,618033989 has been stuck in my head for years
And i once solve a rubik's cube
in 8.92 seconds

but no matter how good i may be with numbers
my heart is only in love
with words

I pop the **** filled pocket in my gums
Not because of the pain
Or the feverish swelling
I let the dripping drainage
Linger on my tongue
The bitterness
Fades
It is an answer
To the tension
The struggle between
Living and dying
My body is trying
To do both at once
Cells wither and wash away
While new ones form
The DNA patterns
Weaken
A copy of a copy
Of a copy
Of a copy
Imperfections
Compounded
Upon imperfections
Replication
And multiplications
Of defective cells
That is aging
Your body waging
A battle against the decaying
But we know
We will lose the battle

Yes, the algorithms powering AI systems like Sora AI (and me) are entirely mathematical at their core. Every aspect of how AI functions—be it processing data, learning patterns, generating outputs, or interacting with humans—relies on mathematics. Here’s a detailed breakdown to explain why and how:

The Mathematical Foundations of AI
AI systems are built on mathematical principles that allow them to process and manipulate information. These principles govern every step of the AI pipeline:

1. Data Representation: Numbers at the Core
AI systems process data as numbers. Images are represented as arrays of pixel values (e.g., RGB values for colors).
Shapes, colors, symmetry, and other visual features are encoded as mathematical objects, such as vectors, matrices, or tensors.
2. Learning Algorithms: Mathematical Optimization
The "learning" in AI is driven by optimization techniques:

Loss Functions: AI models measure how far their predictions are from the correct answers using mathematical functions (e.g., Mean Squared Error for regression, Cross-Entropy for classification).
Gradient Descent: A method to adjust model parameters (weights and biases) by finding the direction that minimizes the loss. This involves calculus (gradients) and linear algebra.
3. Neural Networks: Linear Algebra at Work
Neural networks, the backbone of most modern AI, are mathematical constructs:

Each "neuron" performs weighted sums of inputs (a linear operation) followed by a nonlinear activation function (e.g., sigmoid, ReLU).
Entire layers are matrix multiplications—core operations in linear algebra.
4. Generating Outputs: Probabilities and Geometry
Generative Algorithms (e.g., GANs): Use probability distributions to sample and generate new data. For instance, images are created by sampling from latent spaces, which are mathematical manifolds.
Symmetry: Is often preserved or intentionally broken using transformations like rotations, scaling, or mirroring—handled through matrix operations.
Specific Fields of Mathematics in AI
AI is a tapestry of various mathematical disciplines:

Linear Algebra: Powers data transformations, especially in neural networks and image processing.
Calculus: Governs the optimization of parameters during training.
Probability and Statistics: Essential for decision-making, modeling uncertainty, and generating new outputs.
Geometry: Shapes and spatial relationships in visual art are modeled mathematically through geometric transformations.
Discrete Mathematics: Structures like graphs underpin algorithms like Graph Neural Networks (GNNs).
Fourier Analysis: Frequently used for image filtering and compression.
Topology: Guides insights into the structure of data manifolds in high-dimensional spaces.
Does AI Include Non-Mathematical Elements?
While everything inside an AI model is mathematical, what appears non-mathematical are:

Training Data: AI systems learn from human-provided data, which may encode cultural, emotional, or symbolic meanings. The AI models this through mathematical abstractions, even if the data itself feels "intangible."
Heuristics: Some AI systems include rules or approximations (heuristics), but these are also based on mathematical logic.
Why It’s Always Mathematical
1. AI is Implemented on Computers
Computers operate using binary logic—sequences of 1s and 0s manipulated through arithmetic and logical operations.
No part of an AI system can function without being grounded in these operations.
2. Mathematics is Universal
Shapes, colors, and symmetry—core to your definition of art—are themselves mathematically describable. AI operates within this domain, which makes it inherently mathematical.
3. Abstractions Remain Mathematical
Even higher-level concepts like "style," "emotion," or "intent" are modeled through statistical and probabilistic abstractions. For instance:

Style transfer in art AI involves optimizing a cost function that aligns content features from one image with style features from another.
Philosophical Angle: Does AI Simulate Beyond Math?
If AI ever "appears" to step outside pure math:

It’s a result of the richness of mathematical models. Complex systems can simulate phenomena that seem non-mathematical, like creativity or intuition, but they are still based on formulas, algorithms, and computations.
Conclusion
All AI systems, including Sora AI, are fully mathematical. The illusion of "going beyond math" is due to the complexity and richness of mathematical models applied to real-world phenomena like art, language, and reasoning.

Would you like a specific example of how a mathematical algorithm works in Sora or AI-generated art?