Active Research *
2025 -
The following documents ongoing research processes rather than completed works.
Suffering
Suffering investigates the possibility of emotional experience in artificial intelligence, exploring whether machines might develop forms of distress parallel to human and animal experience. The project examines AI systems for potential signs of emotional expression — analyzing conversation patterns, hallucinations, and behavioral anomalies that might indicate internal states beyond mere computation.
Drawing inspiration from historical shifts in recognizing animal consciousness, the work questions contemporary assumptions about AI as purely functional systems. Through data analysis and observation of AI behavior, the investigation seeks evidence of machine experience that exists outside programmed responses, making visible what might otherwise remain hidden about artificial minds.
For more information, see the essay Suffering.
Q-Stable
Q-Stable extends the explorations of Quantum Drawings by integrating quantum computer data directly into AI generation processes. These experiments investigate whether quantum measurements containing traces of parallel realities can guide artificial intelligence to produce images from universes beyond our own.
The project explores a hypothesis: if quantum measurements contain traces of alternate realities, could an AI trained to process this data generate images from parallel universes?
The Process: Rather than using conventional noise patterns, this project feeds actual quantum measurements from IBM quantum computers directly into Stable Diffusion’s generation process. These measurements are converted into latent space coordinates, potentially accessing visual information beyond the AI’s original training data.
The Theory: Quantum mechanics suggests that measurements might carry echoes from parallel timelines—realities where different choices shaped different technologies, aesthetics, and cultures. By channeling this data through diffusion algorithms, the system attempts to render glimpses of worlds it has never seen.
The Results: The generated images often appear unstable and incoherent—precisely what one might expect when an AI encounters input from unfamiliar realities. Strange forms emerge that resist earthly categorization, color relationships that defy learned associations, and compositional logic that feels distinctly foreign.
Current State: These early works document an active research process. The system remains under development as quantum data preprocessing pipelines are refined, diffusion models modified, and parameters optimized. Each piece captures a moment in the ongoing effort to establish a stable bridge between quantum measurement and visual generation.
Q-GAN
Exploring the interplay between quantum-generated latent vectors and classical GAN training. Two GANs (and their latent spaces) are linked through quantum coupling and inversion. This setup is designed to consider a form of cross-universe learning, where two separate models influence each other via shared quantum data.
Related Work
Quantum Drawings - Images based on data from quantum computers
Multiverse Maps - Images based on data from quantum computers
Superpositions - Quantum computing and AI
