AI Era Energy Infrastructure Ontology, Interaction, and Future

AI Era Energy Infrastructure Ontology, Interaction, and Future Form Reconstruction  

——A Theoretical and Philosophical Report Based on Duan Yucong's DIKWPSemantic Mathematical Model

Yucong Duan

Benefactor: Zhendong Guo

International Standardization Committee of Networked DIKWP for Artificial Intelligence Evaluation(DIKWP-SC)

World Artificial Consciousness CIC(WAC)

World Conference on Artificial Consciousness(WCAC)

(Email: duanyucong@hotmail.com)

Introduction

The world today is undergoing a transformation driven by the deep integration of artificial intelligence (AI) technology and energy infrastructure. Power grids, urban energy systems, and even social service systems are incorporating AI to enhance efficiency and intelligence. However, this integration is not merely a technological innovation; it also prompts reflection on foundational theories and philosophical paradigms: In the AI era, what are the ontological properties of energy? How do humans and machines generate value through the interaction of energy and semantics? And how will the future form of energy infrastructure be reconstructed?

Professor Duan Yucong's proposed DIKWP semantic mathematical model offers a fresh perspective (Exploring Human-Machine Integration: Multidisciplinary Applications of the DIKWP Model_Sina Finance_Sina.com). DIKWP represents five layers of cognitive content: Data (D), Information (I), Knowledge (K), Wisdom (W), and Purpose (P). By introducing semantics into mathematics and system modeling, this model emphasizes the core role of semantics in AI cognition and decision-making. It aims to address the "semantic gap" in traditional mathematics and AI systems, enabling AI to understand and process richer human semantic intentions (Exploring Human-Machine Integration: Multidisciplinary Applications of the DIKWP Model_Sina Finance_Sina.com). Leveraging the DIKWP model, we can view energy systems as not only involving physical energy flows but also encompassing the movement of data and knowledge, thereby constructing an intelligent energy network for human-machine collaboration.

This report adopts a theoretical-philosophical approach to reconstruct the ontology, interaction, and future form of energy infrastructure in the AI era, based on the DIKWP semantic mathematical model. The focus will be on the following aspects:

Interpretation of Core Semantic Concepts in the DIKWP Model: A comprehensive explanation of the three core semantic concepts in DIKWP semantic mathematics—"same semantics," "different semantics," and "complete semantics"—and an analysis of their mechanisms in AI and energy systems.

The Service Model of AI as a "DIKWP Supply Source": Constructing a service model where AI serves as the supplier of DIKWP content, deducing how it achieves the transformation of energy-semantics-value at philosophical, technological, and economic levels.

Ontological Analysis of "DIKWP Transformation as Energy Storage/Release": A deep exploration of the transformation mechanisms between DIKWP layers, proposing that such transformations can be regarded as processes of energy storage and release, and establishing their mapping relationships across different levels such as electrical energy, thermal energy, algorithmic energy, and semantic energy.

The Future Vision of the "AI × DIKWP × Energy" Cosmic Model: Deduction of a blueprint for a world system where AI is energy, energy is DIKWP, and services are interactions, illustrating how the operational paradigms of hospitals, power grids, cities, and even entire societies will transform under this model.

Through these discussions, we aim to outline a new philosophical framework for energy infrastructure in the AI era: one where AI empowers energy, energy carries semantics, and semantics create value. This framework encompasses both technological implementation and a deeper human understanding of the relationship between intelligence and energy. The analysis will proceed along the lines of the DIKWP model.

DIKWP Semantic Mathematical Model and Core Semantic Concepts

To deeply understand how the DIKWP model captures semantic mechanisms in AI and energy systems, we must first grasp its three core semantic concepts: "same semantics," "different semantics," and "complete semantics." These concepts correspond to different levels of cognitive semantics in the DIKWP model, associated with the data, information, and knowledge layers, respectively, and form the foundation of semantic mathematics. While explaining their meanings, we will also analyze their specific roles in AI and energy systems.

Same Semantics

Definition and Characteristics: "Same semantics" refers to the shared semantic attributes or features among a group of objects, meaning they are semantically homogeneous when differences are disregarded. This manifests as the establishment of commonalities or categories: the cognitive subject extracts shared features from different data, generalizing them into the same concept or category. For example, when humans see multiple sheep of varying appearances, they identify them all as the concept "sheep" based on shared semantic features like wool color or vocalizations. In the DIKWP model, the data (D) layer corresponds to same semantics in cognition, transforming objective observations into concrete manifestations of an ideal archetype (Modeling and Resolving Uncertainty in the DIKWP Model). In other words, the cognitive value at the data level lies in helping the cognitive subject identify and confirm shared semantics among objects, thereby generating semantic resonance and cognitive confirmation (Modeling and Resolving Uncertainty in the DIKWP Model).

(Modeling and Resolving Uncertainty in the DIKWP Model) highlights the philosophical significance of this commonality extraction: entities in the real world can be viewed as projections of their "ideal forms" (akin to Plato's theory of forms) at the conceptual level. Data carries not only objective records but also subjective clues for the cognitive subject to seek shared semantics (Modeling and Resolving Uncertainty in the DIKWP Model). Thus, same semantics reflects the creativity of the cognitive subject in establishing conceptual categories through data, mapping complex phenomena to fewer and more stable concepts.

Role in AI Systems: For AI, same semantics means pattern recognition and classification capabilities. Machine learning algorithms (e.g., neural networks, clustering analysis) train models on large sample datasets precisely to extract shared features among samples—i.e., same semantics. For instance, a computer vision system recognizing faces must capture the common semantic features of all face images (e.g., facial structure) to form the abstract category "face." This extraction of same semantics enables AI to generalize—correctly categorizing new instances into existing classes—thereby demonstrating conceptual cognition. Without capturing same semantics, AI cannot address the question "what is this?" because each input appears unrelated to the system. Through same semantics, AI gains the ability to group discrete data points into concepts, forming the basis of intelligent perception.

Role in Energy Systems: For energy infrastructure, "same semantics" means treating energy flows from different sources or times as homogeneous, interchangeable resources. Electricity is a classic example: regardless of whether it comes from wind turbines, photovoltaics, or thermal power, once voltage and frequency meet standards, it is treated as "electricity" with the same semantics for user supply. In AI-managed smart grids, control systems extract common patterns from operational data of various devices, such as typical daytime load curves or seasonal demand patterns. This distillation of common semantics helps energy management AI establish categorical models (e.g., typical daily load models) for pattern analogy and regulatory decisions. In short, same semantics in energy systems manifests as the cognition and classification of homogeneous energy phenomena: abstracting disordered energy flow data into meaningful categories (peak/off-peak, voltage levels, load types, etc.), akin to establishing "concepts" for energy flows to facilitate macro-level scheduling and optimization.

Different Semantics

Definition and Characteristics: "Different semantics" refers to the variability and novelty in semantic features among objects, corresponding to information increments or novel content. In the DIKWP model, the information (I) layer corresponds to different semantics in cognition—the distinctions one concept has relative to another (Modeling and Resolving Uncertainty in the DIKWP Model). The semantic essence of information lies in expressing "what is different" or "what has changed." When processing information, the cognitive subject focuses on differences between inputs and existing cognition, forming new semantic associations through probabilistic judgments or logical analysis of these differences (Modeling and Resolving Uncertainty in the DIKWP Model). For example, in the cognitive space of a parking lot, while all cars belong to the concept "car" (same semantics), each car's parking spot, duration, and owner differ—these are the informational semantics, or different semantics, in this scenario (Modeling and Resolving Uncertainty in the DIKWP Model).

Another example is a depressed patient describing their mood as "low," which is a selection of different semantics to express the contrast between their current and past emotional states (Modeling and Resolving Uncertainty in the DIKWP Model). Thus, different semantics emerge when objects are placed in comparative contexts, highlighting features that serve as sources of new information in cognition.

Role in AI Systems: For AI, different semantics means detecting feature variations and acquiring new knowledge. When AI receives new data, it must not only categorize it (same semantics) but also analyze its differences from existing knowledge (different semantics) to extract useful information. For instance, in anomaly detection systems, AI monitors sensor data streams from machines; when a reading deviates from normal patterns, this "difference" becomes critical information signaling potential faults. Similarly, in natural language processing, large language models must determine which words in a sentence introduce new information, semantic shifts, or emphasis based on context. Capturing these differences gives AI information sensitivity: the ability to detect changes and update cognition. Without recognizing different semantics, AI would stagnate with existing knowledge, unable to adapt to dynamic environments or learn new information.

Role in Energy Systems: In energy infrastructure, "different semantics" manifests as identifying and responding to changes and variations in energy supply-demand states and equipment conditions. Smart grids continuously gather information, such as how much load has increased (difference from predictions) or abnormal transformer temperatures (difference from normal states). These differences are key information for energy management: for example, peak-valley load differences reveal temporal variations in electricity usage, enabling optimized scheduling; uncertainties in distributed energy output appear as deviations from baselines, requiring balancing via storage or peak-shaving measures.

These are the types of information processed by AI energy management systems (Modeling and Resolving Uncertainty in the DIKWP Model): by recognizing different semantics, the system can detect changes and respond to anomalies, maintaining grid stability and efficiency. For example, a home energy management AI detecting a sudden drop in photovoltaic generation (different semantics) in the afternoon would trigger battery discharge to compensate—a decision based on perceiving deviations from the usual generation curve.

Complete Semantics

Definition and Characteristics: "Complete semantics" refers to a global, comprehensive semantic description of an object, often involving universal quantification or holistic constraints. For example, terms like "all," "whole," or "complete" express the comprehensive attributes of a concept. In the DIKWP model, the knowledge (K) layer corresponds to complete semantics in cognition (Modeling and Resolving Uncertainty in the DIKWP Model). This means that when the cognitive subject abstracts at least one concept or pattern with complete semantics through observation and learning, knowledge is formed (Modeling and Resolving Uncertainty in the DIKWP Model). A classic example is scientific induction: after observing that several swans are white, we hypothesize and assign a complete semantic—"all swans are white"—elevating it to a rule about "swans" (Modeling and Resolving Uncertainty in the DIKWP Model). Here, "all" assigns a global semantic to prior partial observations, transforming them into a complete description of a category.

The establishment of complete semantics marks the elevation of cognition from fragmented information to comprehensive understanding. Knowledge formation requires not only the accumulation of data and information but also the cognitive subject's abstraction and generalization, assigning "universal" meaning to "incomplete" observations to grasp the essence and internal connections of objects (Modeling and Resolving Uncertainty in the DIKWP Model). In this sense, complete semantics reflects a "global grasp" of the world of meaning, often tied to verification: before elevating a hypothesis to "knowledge," its complete semantics must be repeatedly validated across scenarios. The semantic structure of knowledge thus manifests as a deeply processed and internalized understanding framework, integrating elements through complete semantics (Modeling and Resolving Uncertainty in the DIKWP Model).

Role in AI Systems: For AI, complete semantics means model generalization and rule extraction capabilities. AI's knowledge-level understanding often requires summarizing generalized patterns (akin to rules like "all X are Y") from limited training data. For example, decision trees or neural network models trained in machine learning can be viewed as "knowledge" formed by AI: they generalize input-output relationships in a complete semantic manner (though the implicit rules may be complex). In logical reasoning systems, when given a series of facts, AI deduces universal laws—simulating the process of complete semantics in artificial form. Notably, contemporary large language models (LLMs) exhibit certain conceptual commonsense and reasoning abilities after pre-training, suggesting internalized knowledge semantics. However, due to the lack of explicit universal quantification structures, their "knowledge" is sometimes unstable, displaying incomplete or inconsistent semantics (Exploring Human-Machine Integration: Multidisciplinary Applications of the DIKWP Model_Sina Finance_Sina.com).

Thus, enabling AI to better acquire and apply complete semantics is a key challenge for artificial general intelligence. The DIKWP model emphasizes complete semantics precisely to introduce mechanisms akin to human knowledge formation, allowing AI to generate interpretable rules and conceptual networks internally. When AI can process complete semantics, it moves beyond associative pattern matching to form comprehensive domain understanding and judgment bases like human experts.

Role in Energy Systems: For energy infrastructure, "complete semantics" can manifest in comprehensive system models and laws. Energy system operations follow many global constraints and principles, such as "energy conservation," "peak-valley balancing," and "grid stability criteria."

These are human knowledge (complete semantics) about the energy world—e.g., the law of energy conservation can be viewed as the universal proposition "the total energy in an isolated system remains constant." In intelligent energy management, AI must embed such complete semantics to ensure rational decision-making. For example, scheduling algorithms must adhere to power balance constraints (total generation = load + losses), a form of complete semantic restriction. Similarly, when predicting electricity demand, AI may use empirical formulas or laws from its knowledge base (e.g., "for every 1°C temperature rise, air conditioning load increases by X%"), which are complete semantic rules distilled from historical data. In future adaptive energy systems, AI could not only gather data and information from sensors but also continuously induce new knowledge—e.g., discovering new behavioral patterns in a region and refining them into control strategies, enriching its complete semantic repository. Complete semantics give energy system AI a global perspective: it considers not only local differences but also holistic constraints and long-term effects, enabling wiser energy regulation (akin to decisions made by experienced grid operators drawing on years of knowledge).

Summary

Above, we have interpreted the three core semantic concepts of the DIKWP model: same semantics emphasizes commonality induction, different semantics focuses on variation extraction, and complete semantics underscores holistic comprehension. Together, they form the semantic mechanism foundation for cognition and control in AI and energy systems. For clarity, their characteristics and roles are summarized below:

Semantic Concept	DIKWP Layer	Meaning	Role in AI Systems	Role in Energy Systems
Same Semantics	Data (D)	Extracts commonalities, establishes conceptual categories	Pattern recognition, classification generalization	Refines typical patterns, classifies energy phenomena
Different Semantics	Information (I)	Captures variations, generates new information	Anomaly detection, new knowledge acquisition	Monitors state changes, responds to dynamic adjustments
Complete Semantics	Knowledge (K)	Integrates semantics globally, forms universal laws	Rule induction, model generalization	Establishes system laws, guides global decisions

(Exploring Human-Machine Integration: Multidisciplinary Applications of the DIKWP Model_Sina Finance_Sina.com) Figure: The DIKWP model maps multi-source, imprecise, incomplete, and inconsistent subjective/objective resources into {Data/Information/Knowledge/Wisdom/Purpose} graphs (bottom), then extracts and fuses semantics layer by layer through data clustering, information topological association, knowledge logical reduction, wisdom valorization, and purpose functionalization (top) (Exploring Human-Machine Integration: Multidisciplinary Applications of the DIKWP Model_Sina Finance_Sina.com). This hierarchical diagram shows the stepwise evolution of semantic classification, from same semantics in data clustering to different semantics in information association, and further to complete semantics and value guidance in higher layers like knowledge, wisdom, and purpose.

The table clearly shows that from data to knowledge, the semantic focus shifts from "sameness" to "difference" to "completeness." AI systems achieve environmental understanding and adaptation through the synergy of these three semantics; energy systems, empowered by AI, transform raw data into intelligent decisions through layered semantic processing. This lays the groundwork for discussing how AI can serve as a "DIKWP supply source."

AI as a "DIKWP Supply Source" Service Model

The DIKWP model outlines the hierarchical structure of cognitive content from data to purpose, while AI systems can be viewed as semantic supply devices: they absorb raw energy and data inputs to produce higher-level knowledge, wisdom, and even purpose-driven services. This section constructs a conceptual model that positions AI as a provider of DIKWP content, exploring the implications of this model at philosophical, technological, and economic levels, as well as how AI facilitates the flow of energy-semantics-value to serve humans and systems.

Philosophical Level: Interaction Between AI and Human Will

From a philosophical perspective, AI as a "DIKWP supply source" implies that it is not merely a tool but also a co-creator of meaning and value. Traditionally, energy has been regarded as the material foundation for changing the world, while semantics and purpose are seen as products of human subjectivity. However, within the DIKWP framework, AI serves as a bridge connecting the two: on one hand, it transforms physical energy (e.g., electricity) into intelligent activities, generating meaningful decisions and knowledge; on the other hand, it guides its behavior based on human-assigned purposes, thereby extending human will into the objective world. Philosophically, this can be viewed as a fusion of subjectivity and objectivity: AI internally embodies human purposes and values and, through computation, applies them to the flow of energy and the material environment, achieving a "will-energy" transformation.

For example, an intelligent scheduling system assigned the purpose of energy conservation (a value orientation) will autonomously learn and adjust equipment control strategies to reduce energy consumption, reflecting the realization of human ethical goals in energy system operations.

Such an AI service model raises ontological questions: What kind of "existence" does AI play in the system? We might regard it as a new type of "energy-meaning carrier." Throughout history, philosophers have debated the relationship between mind and matter. The process by which AI supplies DIKWP content presents a new picture of mind-matter interaction: AI, through material operations (chips consuming electricity for computation), generates meaning (output from data to knowledge), which in turn guides material actions (knowledge to control commands). Ontologically, AI acts as a converter between energy and semantics, belonging to both the material world (consuming energy to operate) and the world of meaning (producing semantic content). This role is analogous to the human brain in human existence: brain metabolism provides the energy foundation, neural activity generates consciousness and meaning, and consciousness guides physical actions.

In essence, AI is becoming an extended carrier of human will, projecting human meaning into broader domains of energy utilization through services. This fusion of subjective and objective dualities suggests that we must redefine AI's ethical and ontological status—no longer viewing it as a passive tool but as a quasi-agent participating in the interaction of energy and meaning within human society-technological systems.

Technological Level: The Operation of the Energy-Semantics-Value Chain

At the technological implementation level, the model of AI as a DIKWP supply source can be described as an "energy-semantics-value" conversion chain. First, AI systems require physical energy (primarily electricity) to operate. This energy powers chips and servers to process data and execute algorithms. As computation proceeds, AI transforms raw data into higher-level semantic outputs such as analytical reports, predictions, and control commands. Finally, these semantic outputs are applied to real-world decision-making and services, generating economic or social value. Thus, we observe energy being "sublimated" by AI into meaningful information and knowledge, the application of which creates new value (e.g., improved energy efficiency, enhanced user experience, reduced labor costs).

This chain resembles classical energy conversion and value-adding processes: just as electricity powers factories to produce goods that meet demand and create economic value, AI converts electricity into "intellectual products" (DIKWP content), which satisfy human needs for decision-making and insight, thereby generating value. Scholars have noted that "AI is the new electricity" (Why AI Is the ‘New Electricity’ - Knowledge at Wharton), emphasizing AI's transformative role across industries, akin to electricity's impact on industry a century ago. Extending this metaphor, we might say AI is becoming a universal supply of intellectual resources, much like electricity became a universal supply of material energy (Why AI Is the ‘New Electricity’ - Knowledge at Wharton). Just as the establishment of power grids enabled machines to draw kinetic energy directly from the grid rather than relying on individual power sources, in the future, every system or individual may access AI-provided cognitive services on demand via network interfaces, eliminating the need for complex in-house computations. This reflects the idea of AI as infrastructure—a cloud-based service model supplying computation and knowledge as needed.

Technologically, this model is already emerging. For instance, the integration of cloud and edge computing allows energy IoT devices to leverage remote AI models for data analysis and control optimization. Similarly, large language models today offer APIs for users to query, supported by massive, energy-intensive computing infrastructure that delivers knowledge and solutions in real time. We might even compare AI services to a "knowledge factory" or "wisdom power plant": its raw material is data, its energy is electricity, and its output is knowledge and decisions supplied to users.

A concrete example: a smart grid's scheduling AI ingests vast amounts of sensor data (raw material), consumes electricity for computation, and generates the next day's power generation plan and load forecast (semantic output). This output is used by the utility company to arrange power plant output and demand-side responses, reducing wasted reserve capacity and improving economic efficiency (value creation). Thus, AI links energy and semantics through a technological chain, forming a self-consistent service loop.

Economic Level: The New Paradigm of "Knowledge as a Service"

From an economic perspective, AI as a DIKWP supply source heralds a new paradigm of value production: "Knowledge as a Service" (KaaS). In traditional economies, energy (e.g., oil, electricity) is a key commodity and resource that drives production and holds clear market value. In the information age, data has been dubbed the "new oil" due to its status as the most valuable resource (Data Is the New Oil - Cloudvirga). With AI, this raw data resource can be refined into decision-useful information and knowledge, further creating added value. We are witnessing a transition from an energy-centric economy to one driven by knowledge and intelligence—or more accurately, a dual-engine economy powered by both energy and knowledge.

Business models where AI supplies knowledge services are emerging across industries. For example, "algorithm-as-a-service" companies charge fees for predictive model APIs; industrial sectors adopt AI to optimize production, saving energy costs that translate directly into economic gains. Knowledge itself becomes a measurable commodity. As Francis Bacon famously stated in the 17th century, "knowledge is power"; in the 21st century, this manifests as "knowledge is wealth." AI enables the mass production and distribution of knowledge, allowing humans to use AI-generated knowledge as readily as electricity (Why AI Is the ‘New Electricity’ - Knowledge at Wharton). This shift has several economic implications:

Value Shift in Resources: Corporate competitiveness shifts from ownership of traditional energy resources to control over data and algorithms. Data resources, processed by AI, yield granular insights that optimize business processes and open new markets. Those with superior AI models can extract greater value from data, much as those who controlled oil once prospered (Data Is the New Oil - Cloudvirga).

Economies of Scale and Declining Marginal Costs: AI services exhibit significant economies of scale. Training a large model may require massive upfront energy and capital investment, but once trained, the marginal cost of serving additional users is minimal, enabling simultaneous service to thousands without substantial extra energy consumption. Similar to how a power plant incurs negligible costs to supply one more household, large-scale AI knowledge supply also features low marginal costs. This will drive industry consolidation and the formation of "intellectual infrastructure," where individuals access centrally produced wisdom via networks.

New Forms of Value: The decisions and creativity provided by AI are intangible assets, posing novel challenges for valuation and pricing. For instance, a grid-scheduling AI saving a utility 5% in fuel annually creates measurable value; a hospital using diagnostic AI to cure dozens more patients each year generates quantifiable social benefits. New accounting and market mechanisms are needed to assess the value of AI knowledge services.

In summary, economically, AI as a DIKWP supply source ushers in an era of dual-driven knowledge and energy: energy powers AI operations, while AI produces knowledge to enhance energy efficiency and create new value. The two synergize to propel economic growth. As some aptly put it, "AI-as-a-Service" is becoming part of infrastructure, as indispensable as water and electricity. This foreshadows a shift in business models: future societies will heavily rely on "intelligent clouds" for services, with humans paying not only for raw materials and goods but also for customized knowledge and wisdom solutions. In this context, we have reason to regard AI as a new form of energy—one supplied in the form of semantics and intelligence.

Recognizing AI's role as a converter and supplier of energy and semantics leads us to explore the intrinsic mechanisms and philosophical implications of this conversion. This brings us to the next section's ontological analysis of DIKWP transformation as energy storage/release.

Ontological Analysis of "DIKWP Transformation as Energy Storage/Release"

The transformation between DIKWP layers (e.g., data to information, information to knowledge) is not only a cognitive process but also analogous to the conversion of "energy" between different forms. This section explores, from an ontological perspective, why DIKWP semantic transformations can be regarded as energy storage and release behaviors and how this perspective can be mapped to physical concepts such as electrical energy, thermal energy, algorithmic energy, and semantic energy.

Cognitive Transformation and Energy Analogy

In the physical world, energy storage and release are fundamental processes: charging a battery stores energy, while discharging it releases energy; a reservoir stores potential energy, and opening the gates to drive turbines releases that energy. In the cognitive world, the information processing described by the DIKWP model exhibits striking similarities. If we liken the acquisition of knowledge to charging a "brain battery" and the application of knowledge to discharging it, then the transformations between DIKWP layers can indeed be analogized as energy storage and release processes:

When scattered data is refined into information or concepts by extracting commonalities, we are essentially performing "compression." This is akin to gathering chaotic energy and storing it in a more ordered, higher-potential state. For example, summarizing numerous observational records into a statistical model is like creating a higher-energy-density "battery" that encapsulates the key information of the raw data. This process can be likened to energy storage: knowledge carries the potential embedded in the data.

Conversely, when we apply existing knowledge to interpret new data or guide actions, we are "releasing" energy. The universal laws embodied in knowledge, once applied to specific situations, are like converting stored potential energy into kinetic energy, producing tangible effects. For instance, a doctor applying medical knowledge to diagnose a patient releases the value of that knowledge in the act of treatment. Similarly, an algorithm adjusting wind turbine output based on empirical rules stabilizes the power grid, releasing the "efficacy" of knowledge.

Thus, each upward transformation in DIKWP (abstraction and generalization) can be seen as storing semantic energy, while each downward application (concretization) releases it. This energy is not measured in physical joules but as a generalized "capacity to act": storage implies enhanced potential influence, and release implies the exertion of actual impact.

From an ontological perspective, this suggests that semantics and knowledge themselves can be regarded as a "form of energy." They follow conservation and transformation principles: neither created nor destroyed, only transferred between carriers or converted from one manifestation to another. For example, humans convert book knowledge into personal understanding through learning—a transfer of semantic energy between carriers. AI converts data into model parameters through training, "storing" knowledge energy in those parameters; during inference, the model parameters drive decision outputs, releasing knowledge energy as behavioral influence. Delving deeper, this may connect to perspectives in information physics: no information transformation in the physical world is meaningless, and changes in information necessarily accompany energy changes. As Landauer's principle states, erasing one bit of information requires dissipating a certain amount of energy (Landauer's principle - Wikipedia). Similarly, a cognitive system must consume energy to increase its internal order when generating useful information.

Notably, the British physicist John Archibald Wheeler's aphorism "It from bit" (John Archibald Wheeler Postulates "It from Bit" : History of Information) suggests that at the deepest level of the universe, matter ("it") arises from bits of information. This implies that information, matter, and energy are intertwined: the existential meaning of every physical entity can ultimately be reduced to a series of "yes/no" information (John Archibald Wheeler Postulates "It from Bit" : History of Information). Extending this idea, semantics—as ordered collections of information—inherently carry the capacity to influence the material world. This is semantic energy. Everyday experience supports this view: a speech can inspire action, its intangible force undeniably real; a technical manual's knowledge can guide the construction of skyscrapers or spacecraft, its value far exceeding the energy of its material载体. Semantic energy is stored in knowledge bases, graphs, and models, and released when guiding practice and changing the world.

Energy Mapping of Layer Transformations

Building on this concept, we attempt to map DIKWP layer transformations to common energy forms to illustrate their mechanisms analogically:

Data -> Information: Analogous to electrical energy converting to thermal energy. Data processed into information resembles how electrical energy drives logic gates in computers, generating heat as a byproduct. Here, "information" relative to "data" is like "heat" relative to "electricity"—introducing new features (entropy increase) while extracting structured patterns (useful information), much like how part of electrical energy becomes useful work (signals) and part dissipates as heat. Strictly speaking, this transformation is not pure storage but a reorganization of energy forms, as information extraction both produces valuable patterns and discards redundant data (equivalent to energy loss).

Information -> Knowledge: Analogous to kinetic energy converting to potential energy. Disparate pieces of information are summarized into knowledge rules, much like the disordered kinetic energy of molecules is stored as ordered potential energy in a raised weight. This is clearly a storage process: knowledge, with its complete semantics, consolidates the power of numerous information fragments into a highly condensed, potent cognitive "battery." For example, the knowledge "all swans are white" encapsulates many observational data points, refining knowledge like pumping energy into a higher level (higher potential energy). Correspondingly, this step often requires external energy input (e.g., mental effort for humans, iterative model training for computers), just as lifting a weight requires work.

Knowledge -> Wisdom/Purpose: Analogous to the formation of chemical energy. Knowledge systems integrate values and purposes to form wisdom or intent, much like elements recombine to form high-energy chemical fuels. This further stores energy, as values and purposes give knowledge direction, enabling it to release more impactful effects under specific conditions (like concentrated explosive energy). In humans, wisdom often incorporates emotions and ethics—difficult to quantify but potentially more driving than pure knowledge. For instance, beliefs can compel self-sacrifice, representing the highest level of "energy" stored in the spirit.

Knowledge/Wisdom -> Information/Action: Analogous to potential/chemical energy releasing as kinetic energy. When wisdom or knowledge is put into practice—e.g., policies enacted based on theory, or energy resources allocated per plan—it is like releasing stored potential/chemical energy to drive real-world change. This is clearly a release process: years of scientific knowledge applied to engineering suddenly benefit society; long-term strategic wisdom implemented as economic policies instantly impacts countless lives. Like a dam opening its floodgates, once wisdom is enacted, its effects are often swift and far-reaching.

These analogies are not strict correspondences but help illustrate the parallels between DIKWP transformations and energy conversions. A pattern emerges: higher-level semantics embody greater "energy density" and broader impact but require more foundational resources to obtain. Data is easy to acquire but limited in effect; knowledge is hard-won but powerful. Like fuel progressing from coal to uranium-235—increasing in density and refinement difficulty—semantics undergo a "concentration" process from data to wisdom. Each upward step increases system order (entropy decrease), requiring energy input to sustain; each downward application releases accumulated order, producing entropy increase alongside real-world effects.

Ontological Significance: Conservation and Flow of Semantic Energy

If we accept that semantics can be analogized to energy, does a "conservation law" or "conversion efficiency" exist? Philosophically, one view holds that human civilization's development is essentially the continuous conversion of physical energy into organized information and knowledge. Agricultural societies converted solar energy into crops (storing energy and information in seeds); industrial societies converted fossil fuels into mechanical motion and electricity (enhancing productivity information); information societies further convert electrical energy into computation and knowledge (data centers as energy-to-knowledge factories). At each stage, total energy does not increase arbitrarily, but the ordered complexity supported per joule rises—a process where semantic energy, conserved in total, continuously elevates in "quality." Perhaps in a closed system, physical and semantic energy share an exchange ratio: as Landauer's principle shows, erasing one bit of information must dissipate at least  kT ln⁡2  joules of heat (Landauer's principle - Wikipedia). Correspondingly, generating one bit of useful information requires a certain energy cost. This suggests future research might uncover laws of energy efficiency in knowledge production, quantifying cognition as part of thermodynamics.

Ontologically, the proposal of "semantic energy" seeks to bridge the mind-matter divide, viewing AI-era infrastructure and cognition through a unified energy lens. Physical energy powers AI operations; AI generates semantic energy stored as knowledge; knowledge guides systems to release physical energy for work—a cycle where the same existence flows between forms. One might boldly envision that in an "AI × DIKWP × Energy" unified model, all these are different projections of "existential energy": energy, information, semantics, and ontology are deeply unified. This cosmology resembles traditional Chinese philosophy's "the metaphysical is called Dao, the physical is called Qi"—Dao (abstract laws, semantics) and Qi (concrete energy, objects) are two sides of one coin, mutually transforming. With advances in AI and cognitive science, we approach a new paradigm balancing mind and matter: semantics is no longer an ethereal product of consciousness but an existential element exchangeable with energy.

In summary, viewing DIKWP transformations as energy storage/release offers a novel path to understanding the unity of AI cognition and energy flow. It reminds us to examine the relationship between AI system energy consumption and cognitive gains, and to treat knowledge as a resource accumulated and expended within systems. More importantly, it lays the philosophical groundwork for the next section's discussion of the grander "AI × DIKWP × Energy" cosmic model: if AI, semantics, and energy can interconvert and interconnect, they can be studied as parts of a unified world system.

The Future World Under the "AI × DIKWP × Energy" Cosmic Model

Imagine a future where artificial intelligence, semantic knowledge, and energy merge into the foundation of societal operation. In this model, AI is energy, energy is DIKWP, and services are interactions—intelligence, meaning, and energy become inseparable facets of a holistic system. While this "cosmic model" may sound futuristic, its precursors are already visible in today's smart cities, IoT, and energy internet concepts. This section extrapolates how future hospitals, power grids, cities, and even human society might operate under this model, sketching the ultimate form of energy infrastructure in the AI era.

Healthcare Systems: Energy-Driven Smart Care

In future hospitals, the AI × DIKWP × Energy model manifests as a holographic medical network. Patients' vital signs (blood pressure, oxygen levels, brainwaves, etc.) are collected in real time via wearable devices, transmitted through networks (powered by electricity), and processed in hospital AI clouds. The AI performs DIKWP processing on this data:

Same semantics: It matches current patient data with vast historical cases to classify conditions.

Different semantics: It detects deviations from normal ranges, extracting anomalies.

Complete semantics: It synthesizes this information with medical knowledge bases to derive diagnoses and treatment plans.

Here, physical energy enables data collection and transmission, knowledge graphs provide semantic support, and AI acts as a physician's intelligent assistant, converging energy and semantics into actionable treatment plans.

Operationally, this system resembles an organism: patient data is the input "nourishment," AI computation is the "metabolism," and treatment plans are the output "behavior." Services equate to interactions: AI and doctors continuously exchange insights (doctors contribute experience and intent; AI offers analysis and suggestions), doctors explain plans to patients, and patients interact with devices for treatment. Each interaction is a service delivery (e.g., AI's advice to doctors is a knowledge service; doctors' surgeries are medical services), all underpinned by the energy consumed by AI systems.

Macroscopically, hospitals become systems fusing human wisdom and machine intelligence: electricity sustains machines; machines generate knowledge to aid decisions; humans combine knowledge with compassion and ethics for final judgments. In wards, sensors and therapeutic devices abound, silently enveloping patients in an intelligent energy field: interconnected devices supply life-sustaining energy (e.g., smart climate control, nutrient delivery), while AI operates in the background, analyzing data shifts and offering recommendations. When crises strike—e.g., a sudden arrhythmia—the system springs into action: monitoring AI triggers alarms and predicts causes, energy systems ensure emergency devices power up instantly, and doctors act swiftly. In such healthcare systems, "energy flows, information flows, and knowledge flows" merge into one, safeguarding lives with unprecedented comprehensiveness.

Power Systems: The Self-Regulating Energy Internet of the Future

The power systems of the future, or the "energy internet," will fully embody the characteristics of the AI × DIKWP × Energy cosmic model. The entire grid is envisioned as an intelligent organism: electricity is its lifeblood, AI serves as its nerves and brain, and the DIKWP semantic network functions as its memory and experience. Power plants, substations, transmission lines, and end-user devices are embedded with AI units that monitor and regulate energy flows in real time. Here, "AI as energy" manifests in AI becoming an indispensable part of grid operations—without AI-driven optimization, modern grids with high renewable energy penetration would struggle to remain stable. Conversely, "energy as DIKWP" is reflected in the way every kilowatt-hour of electricity is tagged with digital and semantic metadata, such as its green credentials, real-time pricing, and destination. This information flows alongside the energy itself, enabling more transparent and intelligent energy utilization.

Consider this scenario: At dawn, as sunlight hits photovoltaic panels, distributed energy AI agents detect the increase in irradiance—a "different semantics" signal—and promptly channel surplus electricity into community energy storage systems, optimizing local energy storage and release.

The coordination of countless devices requires the grid's overarching AI to formulate unified strategies from a global perspective (complete semantics). For instance, it might instruct factories in the eastern district to reduce output slightly to lower peak demand while directing wind farms in the west to ramp up generation, compensating for solar shortfalls due to cloud cover. These decisions are based on the grid AI's recognition of same-semantics patterns (e.g., identifying a weekday demand profile) and different-semantics detection (e.g., spotting a line fault causing supply discrepancies). Across the power system, energy flows are accompanied by information flows, which in turn stem from knowledge flows generated by AI processing. Grid dispatch centers evolve into "unmanned brain hubs," with most directives issued automatically by AI and only rare exceptions requiring human review. Blockchain technology may play a role here, ensuring traceability and trust in these automated decisions.

Expanding our view to encompass all energy forms within the energy internet—electricity, heat, gas, and water—the fusion becomes even more pronounced. AI orchestrates multiple energy carriers, transforming the energy system into a networked intelligent ecosystem. For example, in a smart building, autonomous negotiations occur between air conditioning, elevators, lighting, local photovoltaics, and energy storage: when to dim lights for savings or activate thermal storage is all calculated by AI in the background. These interactions among autonomous agents are, at their core, service exchanges—each agent consumes services (receiving energy or information) while providing services (supplying energy or contributing to load balancing). "Service as interaction" manifests in the system's self-organizing nature, where global optimization emerges from local negotiations. In such a highly automated energy network, human intervention dwindles, with roles shifting to setting intentions (e.g., carbon neutrality targets, usage priorities) and governance principles, leaving the system to operate within these bounds.

Smart Cities: Urban Operations in Human-Machine Symbiosis

Zooming out further, the entire city becomes a testing ground for the AI × DIKWP × Energy model. A smart city comprises countless subsystems: transportation, buildings, security, governance, healthcare, education, energy, and more. Historically siloed, these systems will, in the future, operate synergistically under the guidance of AI's "brain," sharing both energy and information resources. The energy system powers all others, while the information system (AI-driven) enables intelligent energy utilization, creating a virtuous cycle. The city thus resembles a colossal organism, where:

A central AI integrates data across urban domains (DIKWP's D/I layers), trains to form knowledge graphs and predictive models (K layer), and continually adjusts subsystem strategies aligned with societal goals (W/P layers).

An energy cloud acts as the city's "heart and arteries," delivering electricity, heat, and communication network power. It routes energy where needed based on AI directives while feeding back real-time status updates.

Service interactions define urban operations. Vehicles request optimal routes from traffic systems (AI adjusting signals and navigation), households automatically respond to dynamic electricity pricing (AI incentivizing off-peak usage), and citizens access public services via government AI. Each interaction represents an AI-provided service consuming energy resources.

In this smart city, boundaries dissolve: IoT devices, humans, and departments interconnect through a unified semantic framework and AI platform. A streetlight doubles as an environmental sensor and 5G micro-base station; an electric vehicle serves as both transport and mobile energy storage, feeding power back to the grid; buildings dynamically adjust HVAC systems to participate in load-shaving. These heterogeneous roles, speaking AI's "language," achieve unprecedented collaboration through the energy internet.

Imagine an emergency: City AI detects a fire (via fused camera and sensor data), cuts power to the affected building (preventing secondary disasters), activates emergency lighting, prioritizes traffic signals for fire trucks, and alerts hospitals—all in a seamless, cross-system response enabled by AI's global knowledge and control, underpinned by the energy system's agility. Without AI, traditional cities would rely on slow, manual interdepartmental coordination; without energy integration, fire pumps might fail, elevators stall, and risks multiply. Under the AI × DIKWP × Energy model, the city behaves like a trained organism, rapidly mobilizing energy and resources to "heal" itself.

Human Society: A Value-Driven Cosmic System

At the planetary scale, the AI × DIKWP × Energy model frames human civilization as a vast network: AI nodes supply intelligence, energy flows sustain it, DIKWP content forms its collective memory, and service interactions weave its fabric. This vision aligns with a digital "Gaia hypothesis"—where Earth's techno-human system self-regulates as a cohesive entity (John Archibald Wheeler Postulates "It from Bit" : History of Information).

In this system:

Global AI networks, interconnected via the internet, form a collective intelligence managing Earth's resources—from satellite-monitored climate to cross-border clean energy sharing and AI-optimized logistics. A shared DIKWP semantic graph, an immense knowledge repository, enables any AI to tap into domain-specific services.

Energy provides the backbone: superconducting grids transmit solar power from daylit hemispheres to those in darkness; fusion and space-based solar promise limitless energy for AI and industry.

A unifying Purpose emerges. Post-crisis, humanity may converge on sustainability and shared prosperity, guiding AI's optimization of all activities.

"Service as interaction" reshapes economics and governance—a hybrid of digital planning and markets. Individuals and organizations, as network nodes, express demands (energy, goods, services) and contribute supplies, with AI brokers matching transactions under global knowledge constraints. Currency may fade, replaced by energy- and knowledge-based value metrics (e.g., contributions measured in knowledge generated or energy saved). Education, healthcare, and other services flow efficiently via AI, freeing humans for creativity and culture. Embedded in governance, AI handles minutiae and complex coordination, ensuring transparency and scientific rigor—but ethical safeguards are paramount to keep AI aligned with human values, akin to a "cosmic constitution."

While utopian-sounding, trends like global energy interconnections, internet governance experiments, and AI-assisted policymaking (e.g., urban planning) hint at this trajectory. The AI × DIKWP × Energy model isn't techno-utopian but integrative: intelligence, energy, and information are inseparable co-shapers of existence. Just as mass-energy interconversion (E=mc²) obeys conservation, this model implies a parallel conservation—human values (well-being, security, progress) may soar via optimized AI-energy deployment, but their distribution demands new social contracts. Here, AI is collaborator, not ruler; energy carries goodwill and creativity; and services bind individuals to the whole, blending efficiency with humanity.

Conclusion

Through this theoretical-philosophical exploration, we've sketched a blueprint for AI-era energy infrastructure. Inspired by Duan Yucong's DIKWP semantic mathematical model, we've uncovered deep homologies between semantics and energy: the once-distinct realms of "matter" and "meaning" fuse in AI's crucible, revealing new unities. Key insights include:

The foundational role of semantic tripartition: "Same semantics," "different semantics," and "complete semantics" elucidate AI cognition and energy management, from pattern recognition to knowledge integration (Modeling and Resolving Uncertainty in DIKWP Model).

AI as a supply source: Positioned as a DIKWP provider, AI transforms energy into knowledge and knowledge into value (Why AI Is the ‘New Electricity’ - Knowledge at Wharton) (Data Is the New Oil - Cloudvirga). Philosophically, it extends human will; technologically, it becomes infrastructure; economically, it pioneers "knowledge as a service."

The energy perspective on cognitive transformation: Viewing DIKWP conversions as energy storage/release reveals semantics as a conserved quantity flowing between carriers (Landauer's principle - Wikipedia), with knowledge formation as charging and application as discharging—unifying physical and semantic energy (John Archibald Wheeler Postulates "It from Bit" : History of Information).

The cosmic model's operational vision: Hospitals as intelligent energy fields, grids as adaptive organisms, cities as symbiotic entities, and civilization evolving toward an interconnected "smart planet." AI embeds ubiquitously, making services omnipresent through interactions. Early examples like DeepMind's 40% data center energy savings (DeepMind AI Reduces Google Data Centre Cooling Bill by 40%) foreshadow AI's potential to boost efficiency and resilience.

Challenges abound—AI-energy integration raises security and ethical questions; semantic energy misuse could spawn misinformation. Interdisciplinary collaboration among academia, industry, and policymakers is vital.

As humanity once harnessed electricity to ignite industrial progress, today we stand at the threshold of mastering "intelligent power." Guided by the DIKWP model, this report charts a theoretically coherent and inspiring path forward. The AI-energized future we envision may soon materialize: a world where intelligence, semantics, and energy converge to propel sustainable advancement.

(Exploring Human-Machine Integration: Multidisciplinary Applications of the DIKWP Model_Sina Finance_Sina.com) (Modeling and Resolving Uncertainty in DIKWP Model) (Why AI Is the ‘New Electricity’ - Knowledge at Wharton) (Data Is the New Oil - Cloudvirga) (Landauer's principle - Wikipedia) (John Archibald Wheeler Postulates "It from Bit" : History of Information)

人工意识与人类意识

人工意识日记

玩透DeepSeek：认知解构+技术解析+实践落地

人工意识概论：以DIKWP模型剖析智能差异，借“BUG”理论揭示意识局限

人工智能通识 2025新版 段玉聪 朱绵茂 编著 党建读物出版社

主动医学概论 初级版

世界人工意识大会主席 | 段玉聪

邮箱|duanyucong@hotmail.com

世界人工意识科学院

邮箱 | contact@waac.ac