This article introduces "Thermodynamic Computing," a proposed new paradigm rooted in the natural processes of self-organization and energy dissipation, contrasting it with mainstream physics' focus on abstract concepts like quantum mechanics and black holes. The author argues for a shift towards understanding "Life itself" as a physical phenomenon, driven by "Thermodynamic Evolution (TE)." TE posits that structures emerge and persist by continuously dissipating free energy, effectively reframing "survival of the fittest" as the stabilization of successful energy-dissipating pathways. This understanding leads to the concept of "thermodynamic bits" (kT-bits), self-configuring computational units that leverage energy flow to spontaneously solve constraint problems. Demonstrations include solutions for 3-Satisfiability and Sudoku, illustrating how matter can self-organize to find solutions. The author's company, Knowm, aims to develop new processors based on these thermodynamic principles.
Thermodynamic Computing: A New Era
Introduction to Thermodynamic Computing: The article begins by proposing "Thermodynamic Computing" as an emerging era in computation, grounded in the universal natural processes of self-organization and the fundamental laws of physics.
Critique of Mainstream Physics: The author challenges the current scientific focus on abstract realms like Quantum Mechanics (QM), Quantum Computing (QC), and black holes, arguing that this pursuit often overlooks the measurable and tangible physics of "Life itself."
The Absurdity of Quantum Computing: QC is presented as attempting to fight quantum decoherence, a fundamental process that gives rise to observable reality, making it an inherently "insane" engineering challenge. Schrödinger's Cat is invoked to highlight what the author views as the absurdity of QM.
Ignoring Life: A key contention is that modern physics absurdly seeks a unified theory by studying phenomena that cannot be directly measured (qubits, black holes), while neglecting the study of life—a process that creates technology and is directly observable.
A New Framework for Physics: Thermodynamic Evolution (TE): TE is introduced as a comprehensive explanation for how matter self-configures to reduce energetic potentials. It describes "symmetry breaking" and the emergence of complex structures.
Reinterpreting "Survival of the Fittest": TE reframes Darwinian evolution's core principle, stating that "Structure that is responsible for more free-energy dissipation is stable because it uses the energy to repair itself." This is illustrated with examples such as mud brick buildings, the organization of companies, and lions finding food.
Pioneers and Similar Ideas: The article acknowledges a long list of thinkers (including Erwin Schrödinger, Rod Swenson, Adrian Bejan, Alex Wissner-Gross, and Jeremy England) who have articulated similar principles regarding systems seeking to minimize potential or maximize entropy.
AHaH Computing and Knowm Synapses (kT-Bits): The author links TE to their work on AHaH Computing, where a "Knowm synapse" (dubbed a "kT-Bit") arises from competitive energy-dissipating pathways, driving learning and maximizing power dissipation. These kT-Bits are seen as self-organizing building blocks found in natural systems like arteries, neurons, and river networks.
A New Branch of Computing: The understanding of thermodynamic evolution is presented as a direct pathway to developing powerful new technologies, particularly in computing.
Functioning of Thermodynamic Bits: A "thermodynamic bit" (kT-bit) is defined as a binary state that mutates probabilistically based on its ability to "harvest" energy by satisfying environmental constraints (e.g., Boolean clauses). Satisfying constraints allows it to repair its state and achieve stability, while failure leads to instability ("flip-flopping").
Solving Complex Problems with kT-Bits: The concept is applied to NP-Complete problems like 3-Satisfiability (dubbed "PhySAT"), where kT-bits spontaneously configure to satisfy constraints, exhibiting self-organized criticality. An animation of a Sudoku solver is also presented as a visual demonstration.
Real-World Demonstrations: Further applications are highlighted, including solutions for the Traveling Salesman Problem, non-convex optimization, and a robotic arm controller, all showcasing how energy and information processing merge to allow matter to self-configure and solve problems.
Knowm's Technological Goal: The author's company, Knowm, aims to build new processors that directly exploit thermodynamic evolution, asserting that the technological capability for this is already present.
Consilience and Future Outlook: The article concludes by noting the independent discovery of similar concepts (e.g., Memcomputing) and expresses a mix of excitement and trepidation regarding the significant power and societal implications of future technologies derived from thermodynamic evolution.