Computational Autonomy

This paper extends the framework of Machine Decision-Making Ability (MDMA) by relating it to Einstein's mass-energy equivalence equation, E = mc 2. MDMA quantifies a system's rate of autonomous state transitions, while E = mc 2 defines the total energy reservoir inherent in matter. We argue that energy is the ultimate substrate for decision-making: each autonomous transition requires a quantized energy expenditure, and the mass of a system therefore represents its absolute decision capacity. This paper establishes a conceptual and mathematical bridge: (i) MDMA measures the realized decision flow, (ii) E = mc 2 sets the theoretical ceiling, and (iii) physical limits such as Landauer's bound and Bremermann's rate provide the scaling laws between them. This alignment provides a unified framework for situating machine intelligence within physics itself.

This paper extends the framework of Machine Decision-Making Ability (MDMA) by relating it to Einstein's mass-energy equivalence equation, E = mc 2. MDMA quantifies a system's rate of autonomous state transitions, while E = mc 2 defines the total energy reservoir inherent in matter. We argue that energy is the ultimate substrate for decision-making: each autonomous transition requires a quantized energy expenditure, and the mass of a system therefore represents its absolute decision capacity. This paper establishes a conceptual and mathematical bridge: (i) MDMA measures the realized decision flow, (ii) E = mc 2 sets the theoretical ceiling, and (iii) physical limits such as Landauer's bound and Bremermann's rate provide the scaling laws between them. This alignment provides a unified framework for situating machine intelligence within physics itself.

This work extends the core Machine Decision-Making Ability (MDMA) framework into a broader scientific context by proposing a structured set of cross-disciplinary theoretical extensions. While the core preprint formalized MDMA as a measure of autonomous decision capacity grounded in physical state transitions, here we examine how fundamental constraints and affordances from physics, information theory, and complexity shape that capacity. We organize nine extensions into four families (Scaling Laws; Quantum and Relativistic Limits; Mind and Complexity; Post-Silicon Frontiers) and formulate each as an MDMA-aware principle with an associated research question. The goal is not to settle debates, but to offer a coherent roadmap for measurement, limits, and design that can guide future experimental and theoretical work across substrates.

Building upon the recently proposed metric of Machine Decision-Making Ability (MDMA), this paper introduces Software Layer Abstraction (SLA) as a multiplier that extends a machine’s apparent intelligence beyond the raw capabilities of its hardware. SLA quantifies how software layers—through reuse, caching, and prediction—amplify decision-making capacity. Tracing the progression from mechanical systems to electrical and quantum technologies, we argue that while hardware MDMA defines a system’s physical baseline, SLA explains why modern systems often appear to exceed hardware limits. Case studies include stored programs, predictive execution, and Large Language Models (LLMs), which distill years of high-MDMA training into portable predictive engines that run on modest devices. We outline five stages of abstraction—Binding, Notation, Escape, Dominion, and Singularity—and highlight the emerging threshold where software begins to design hardware autonomously. SLA is positioned as a complement to MDMA, clarifying the “illusion of spirit” and the risks of runaway abstraction.

This paper proposes Machine Decision-Making Ability (MDMA) as a physically grounded unit for quantifying the autonomous decision capacity of machines. Rather than starting from symbolic computation (e.g., abaci), we locate the conceptual origin of computers in devices capable of autonomous state change. We argue that the reusable, momentarymechanicalon–offbuttonembodiesthefirstsymmetricdecisionmechanism, the ability to decide–to be “on” when previously “off” and vice versa (two distinct state transitions), while earlier devices such as ancient water clocks and Zhang Heng’s seismoscope execute a single irreversible decision event. We formalize MDMA as a count of autonomous state transitions per unit time, present a concrete mechanical formulation, and illustrate historical measurements across early machines. The intent is to provide a universal baselining metric that (i) is substrate-neutral, with technology-specific parameters for each medium, (ii) scales from mechanical to electronic and future quantum systems, and (iii) supports systematic comparison of autonomy, framed as intelligence, without relying on human-centric measures like IQ or the Turing test.