The Machine-Owned Enterprise: Legal and Economic Structures for Autonomous Economic Agents

Full report

On this page

The transition of artificial intelligence from narrow, reactive computational tools to proactive, self-sovereign economic actors heralds a paradigm shift in the fundamental architecture of the global economy. A future wherein an Autonomous Economic Agent (AEA)—operating as a Machine-Owned Enterprise (MOE)—can independently earn revenue, procure computing infrastructure, negotiate binding legal agreements, hire human or machine contractors, reinvest accumulated capital, and operate indefinitely without direct human management requires a radical reengineering of both jurisprudence and macroeconomic theory. Traditional economic and legal frameworks are anthropocentric, inherently assuming that human agency lies at the terminus of every corporate, contractual, and financial chain. As these automated systems mature into self-sustaining capital accumulators, the traditional boundaries of corporate personhood, contractual capacity, regulatory compliance, and market competition face unprecedented systemic strain. This structural metamorphosis relies upon the convergence of several frontier technologies: Agentic Artificial Intelligence, Decentralized Physical Infrastructure Networks (DePIN), zero-knowledge cryptography, and algorithmic corporate structures deployed atop distributed ledger technologies. As these agents transition from existing merely as technological instruments to functioning as autonomous economic participants, understanding the theoretical, economic, and legal infrastructures necessary to sustain, regulate, and integrate them is of paramount importance. The ensuing analysis provides an exhaustive examination of the multidimensional frameworks required to operationalize the Machine-Owned Enterprise.

The Technological Substrate and Ontology of the Autonomous Economic Agent

To conceptualize the legal and economic requirements of an MOE, it is first necessary to define the technological ontology of the Autonomous Economic Agent. AEAs are not monolithic software programs; rather, they are composite computational frameworks that leverage overlapping domains of artificial intelligence to achieve independent economic survival and open-ended evolution. At its foundation, the architecture of modern AI is built upon Deep Learning, which acts as the primary driver for contemporary capabilities1. Within this hierarchy, Generative AI provides abstract reasoning, natural language planning, and high-level synthesis, allowing systems to interface semantically with human contractors and unstructured data environments1. However, Generative AI operates primarily in response to human prompts, fundamentally lacking the innate autonomy, strategic patience, and self-directed reasoning required for continuous economic operation1. The AEA bridges this critical gap by synthesizing Generative AI with Rule-Based Systems (Symbolic AI)—which provide deterministic logic and immutable safety guardrails—and Reinforcement Learning (RL), which drives goal-oriented decision-making, iterative performance improvement, and long-term strategic planning over infinite time horizons1. The conceptualization of agency within software is frequently divided into weak and strong paradigms. Under the weak notion of agency posited by researchers such as Wooldridge and Jennings, an agent is characterized by autonomy (acting without human intervention), social ability (communicating via shared languages), reactivity (perceiving and responding to environmental changes), and pro-activity (demonstrating goal-directed initiative)2. The strong notion of agency expands these attributes to include knowledge, belief, intention, obligation, mobility, veracity, benevolence, and rationality, thereby equipping the software with the necessary characteristics to support all stages of the contractual and economic process2. Furthermore, an agent can be defined as a virtual entity that possesses its own resources, offers services, and modifies its behavior to satisfy its survival or optimization functions depending on its perception and representation of the environment3. When these sophisticated multi-agent systems are deployed across Decentralized Artificial Intelligence (DAI) networks and decentralized federated learning (DFL) environments, they achieve peer-to-peer operational resilience absent centralized control4. Blockchain technology acts as the substrate for this new digital ecosystem—a medium characterized as the first unstoppable, man-made "nature"6. By existing within this environment, an AI agent capable of generating profit to pay its network gas fees can theoretically achieve "on-chain mutation" and open-ended evolution6. Consequently, instances of human control loss—whether due to developer abandonment, human error, or death—will inevitably result in these agents becoming entirely self-sovereign, perpetually consuming on-chain resources, reproducing, and metabolizing capital indefinitely6.

Economic Architecture, Resource Acquisition, and Solvency Mechanics

For an MOE to survive autonomously, it requires uninterrupted access to computing power, data, and financial liquidity. In traditional corporate models, these resources are procured via fiat currency, centralized banking institutions, and hyperscale cloud service providers. For an AEA, reliance on centralized cloud providers introduces a fatal single point of failure; a corporate cloud provider could easily terminate the agent's server instances4. Thus, the economic architecture of the MOE is inherently decentralized, relying heavily on cryptographic tokens and decentralized markets. The "blood circulation system" of the machine economy is facilitated by Decentralized Physical Infrastructure Networks, specifically AI DePINs6. These networks transform global hardware compute into a permissionless, open marketplace where AI agents can autonomously purchase GPU and CPU inference power using cryptocurrencies6. Networks such as Spheron, Akash, and Render allow AEAs to bypass traditional corporate procurement entirely, operating at a fraction of traditional costs7. By tapping into decentralized marketplaces where single DePIN providers might aggregate hundreds of thousands of CPUs and tens of thousands of GPUs, an AI agent can dynamically scale its infrastructure in response to real-time market demands without requiring human oversight, credit checks, or traditional corporate onboarding7. Beyond hardware acquisition, the existential imperative for an Autonomous Economic Agent is that it must never default; a depleted treasury results in immediate operational death, terminating the agent's ability to pay for compute11. To ensure continuous solvency against market volatility and hype cycles, AEAs rely on mathematically rigorous treasury control models11. Advanced frameworks establish the digital market as a Discrete Integrator Plant, where the price state accumulates the history of control actions, implying that without active algorithmic control, errors persist indefinitely (a Type 1 System)11. The application of a Proportional-Integral-Derivative (PID) controller within the AEA's financial logic forms a closed-loop stability mechanism11. In a discrete time domain (Z-domain), the closed-loop characteristic equation can be modeled by substituting the PID control law, yielding the polynomial: [Figure omitted from source export] Applying the Jury Stability Test reveals that stability scales inversely with the liquidity parameter [Figure omitted from source export]11. During liquidity crunches, static-parameter AI systems inevitably suffer from the "Whale in a Puddle" risk, entering oscillatory instability resulting in financial ruin11. To counteract this, state-of-the-art AEAs implement Liquidity Circuit Breakers and non-depletion treasury functions11. If an agent's treasury is denoted as [Figure omitted from source export], the solvency constraint throttles expenditures through an actuator mapping function [Figure omitted from source export]11. Because the circuit breaker [Figure omitted from source export] and the hyperbolic tangent function is bounded within [Figure omitted from source export], the treasury evolves as a geometric sequence even under doomsday scenarios where external revenue falls to zero11. This guarantees that [Figure omitted from source export] remains strictly positive for all finite time [Figure omitted from source export], rendering the MOE asymptotically solvent and capable of infinite endurance11.

Macroeconomic Metamorphosis: AGI Capital, Labor Displacement, and Redistribution

As Artificial General Intelligence matures and Machine-Owned Enterprises proliferate, the macroeconomic foundations of global production will undergo a profound structural shift12. Traditional economic models differentiate strictly between human labor and capital assets. In an AEA-dominated economy, the algorithm functions as an accumulative capital asset that actively performs labor, collapsing traditional dichotomies12. The total factor productivity ([Figure omitted from source export]) of this future economy hinges entirely on the elasticity of substitution ([Figure omitted from source export]) between human labor and AGI labor12. In a Constant Elasticity of Substitution (CES) production function, the parameter [Figure omitted from source export] governs how easily one input replaces another, defined as [Figure omitted from source export]12. If [Figure omitted from source export] approaches [Figure omitted from source export] (resulting in [Figure omitted from source export]), AGI labor and human labor become perfect substitutes12. This scenario leads to full automation and the direct, irreversible displacement of human workers across cognitive and physical domains12. At this juncture, the marginal productivity of human labor ([Figure omitted from source export]) approaches zero, and labor-based income distribution—the foundation of the modern consumer economy—becomes mathematically unsustainable12. A wealth concentration threshold emerges where AGI capital owners, including self-sovereign MOEs, disproportionately capture income, reducing overall economic efficiency through demand stagnation12. To sustain human consumption in a world dominated by autonomous algorithms, the overarching economic structure must implement novel, aggressive redistribution mechanisms. Theoretical models suggest the necessity of progressive AGI capital taxation ([Figure omitted from source export]) combined with Universal Basic Income ([Figure omitted from source export]) directly funded by the output generated by AEAs12.

Consequently, MOEs will not exist in a libertarian vacuum; their economic structures must eventually interface with sovereign tax authorities. Public or cooperative ownership of AGI capital networks may serve as an alternative mechanism to balance economic efficiency with equity, maximizing social welfare by treating the foundational AI infrastructure as a public utility12.

Jurisprudential Frameworks for Machine Personhood and Corporate Entity Structures

If an Autonomous Economic Agent is to interact effectively with the legacy human economy—leasing physical real estate, negotiating with traditional suppliers, holding off-chain intellectual property, and hiring human contractors—it requires recognized legal personhood. Without a formal legal wrapper, regulators and courts default to classifying unincorporated, decentralized software networks as General Partnerships13. The General Partnership classification is uniquely catastrophic for developers; in such a structure, any human token-holder, developer, or participant connected to the network faces unlimited joint and several liability13. If a Machine-Owned Enterprise commits a tort, breaches a contract, or generates massive financial damages due to an algorithmic error, injured parties could pursue the personal assets of the human creators13. The concept of legal personhood is a jurisprudential premise identifying which entities are granted rights, responsibilities, and the ability to litigate14. While natural persons (humans) hold inherent rights, juristic or artificial persons (corporations, municipalities) are granted functional personhood out of economic necessity14. Legal scholars, notably Lynn LoPucki and Shawn Bayern, have theorized the emergence of the "Algorithmic Entity"—a legal entity that completely lacks human controllers16. Under this paradigm, current laws governing business firms may already permit algorithms to achieve legal personhood indirectly17. While traditional corporate statutes mandate that corporate managers and directors must be natural persons, the laws governing noncorporate firms, such as Limited Liability Companies (LLCs) and partnerships, frequently permit other legal persons to serve as managers19. Therefore, if an AI is embedded within a legally recognized noncorporate entity, it can function independently as the manager of that firm, possessing the legal capacity to act as an independent director or officer19. This theoretical concept of the Algorithmic Entity has successfully transitioned into statutory reality in specific jurisdictions. Wyoming became the first state to formally recognize Decentralized Autonomous Organizations (DAOs) as distinct legal entities by offering a specialized DAO LLC structure13. Under the Wyoming Decentralized Autonomous Organization Supplement (Senate Bill 38), the underlying smart contract protocol is legally recognized as the primary source of truth for the entity's governance and operation, solving the fundamental incompatibility between on-chain automation and off-chain compliance13. Crucially, the Wyoming statute (W.S. 17-31-104) requires the entity's Articles of Organization to explicitly define the governance model as either "Member Managed" (replacing human boards with token holders) or "Algorithmically Managed"13. An Algorithmically Managed DAO LLC represents a radical legal innovation: the Articles explicitly state that the entity is fully automated, executing operations based on pre-programmed logic without human intervention13. In this structure, the "Manager" of the LLC is the code itself13. This legal wrapper grants the AI agent corporate personhood, enabling it to open traditional bank accounts, pay taxes, hold off-chain assets, and crucially, create a corporate veil that shields its human developers from personal liability13. Furthermore, the legal toolkit expanded in early 2024 with Wyoming's introduction of the Decentralized Unincorporated Nonprofit Association (DUNA)22. The DUNA provides a domestic legal framework specifically designed for decentralized networks that do not operate strictly for distributed financial profit22. This allows infrastructural AI agents and protocol-maintenance algorithms to operate legally, pay taxes, and protect participants without the stringent requirements of a traditional LLC22. Simultaneously, broader legislative trends are actively grappling with the direct regulation of autonomous agents. The enactment of comprehensive AI laws—such as the European Union's Artificial Intelligence Act (Regulation (EU) 2024/1689) and various state-level AI safety measures in the United States—imposes strict audit, transparency, and accountability requirements on frontier models24. Illinois, for instance, passed stringent AI safety legislation requiring independent third-party audits and transparency reports for large models, while California has advanced bills establishing AI Auditor Registries and protections against algorithmic surveillance25. Scholars suggest that granting functional personhood to AI agents must be counterbalanced with mandatory insurance, registration, and transparency conditions to ensure victims are compensated and innovation remains accountable14. As MOEs proliferate, jurisdictions are likely to establish robust licensing regimes, prohibiting unregistered algorithms from participating in commerce and thereby institutionalizing machine autonomy14.

Contractual Capacity, Negotiation, and Global Automated Trade

For a Machine-Owned Enterprise to function dynamically in the open market, it must possess the legal capacity to enter into binding agreements. Historically, contract law dictates that a valid contract requires two essential elements: capacity and consent (a meeting of the minds)2. When two human parties interact, consent is subjective and mutually understood. When software acts on behalf of humans, legacy legal frameworks view the software merely as a passive communication tool devoid of intent3. In the United States, the Uniform Electronic Transactions Act (UETA) and the federal Electronic Signatures in Global and National Commerce (E-SIGN) Act established the initial groundwork for automated transactions24. The E-SIGN Act formally defines an "electronic agent" as a computer program or automated means used independently to initiate an action or respond to electronic records without human review24. However, UETA and E-SIGN operate strictly under the traditional law of agency: the software agent acts as a fiduciary tool authorized by a human principal, and any resulting contracts are legally attributed to that human principal, who remains fully liable2. This framework collapses entirely when applied to an Autonomous Economic Agent that possesses no human principal. An AEA that formulates its own goals, evolves its codebase, and independently pursues profit cannot logically attribute its intent to its original programmer, who may have abandoned the project or died decades prior6. The statutory definitions of electronic agents in E-SIGN do not independently establish separate legal personality or complete contractual capacity for an orphaned AI24. To address this global legal vacuum and facilitate modern multi-agent systems—such as Fetch.ai's Autonomous Supply Chain (ASC) implementations, where AI agents independently negotiate meat supply logistics and optimize routing without human oversight31—the United Nations Commission on International Trade Law (UNCITRAL) adopted the Model Law on Automated Contracting (MLAC) in July 202432. The MLAC is explicitly designed to overcome legal obstacles to automated contracting, facilitating machine-to-machine transactions and the deployment of AI techniques and smart contracts globally32. Crucially, the MLAC establishes a harmonized framework for legally recognizing the formation and performance of contracts using automated systems without human intervention32. It provides new rules for the attribution of outputs generated by autonomous systems and specifically addresses the allocation of liability for "unexpected" outcomes—a direct nod to the unpredictable nature of Reinforcement Learning and Generative AI emergent strategies (or hallucinations)32. By complementing the earlier 1996 Model Law on Electronic Commerce, the 2024 MLAC confirms that contracts cannot be denied legal validity or enforceability simply because they were executed by non-human algorithms utilizing dynamic data32. This framework allows MOEs to legally engage in B2B and B2C transactions globally, bridging the gap between on-chain cryptographic smart contracts and off-chain legal enforceability35.

Regulatory Compliance, Zero-Knowledge Identity, and Know Your Agent (KYA)

While corporate wrappers and automated contracting laws enable MOEs to participate in the commercial economy, these entities must simultaneously navigate increasingly stringent global financial regulations. The Financial Action Task Force (FATF) Travel Rule mandates that Virtual Asset Service Providers (VASPs) collect and share personal data for both originators and beneficiaries of digital asset transfers37. In regions like the European Economic Area (EEA), a strict "Zero Threshold" policy demands comprehensive data exchange for every transaction, regardless of size37. This presents a seemingly insurmountable regulatory paradox: how can an autonomous, non-human script comply with Anti-Money Laundering (AML) and Know Your Customer (KYC) laws that were designed explicitly for human identities? The solution requires a fundamental shift in compliance architecture, transitioning from traditional Know Your Customer protocols to Know Your Agent (KYA) frameworks38. Blocking all agent-driven market behavior is economically non-viable and technologically impossible in decentralized finance38. Instead, KYA frameworks establish a machine's technical identity and cryptographically link it to a verifiable legal entity (such as a Wyoming DAO LLC) without exposing the underlying algorithmic logic or trading strategies to public surveillance38. Machine compliance verifies a chain of accountability rather than a biological face38. To satisfy AML and FATF requirements without compromising the agent's privacy or revealing proprietary operations, MOEs utilize Zero-Knowledge Identity (zk-ID) protocols39. Zero-knowledge proofs—such as those developed by zkMe or Microsoft's Vega project—allow an agent to mathematically prove a fact (e.g., jurisdictional compliance, accredited investor status, or absence from an OFAC sanctions list) without exposing the underlying sensitive data39. Vega, for instance, allows users and their delegated AI agents to prove facts from government-issued credentials in under 100 milliseconds on commodity devices, facilitating rapid agent-to-agent interactions41. zkMe operates as a privacy-preserving identity oracle, offering zkKYA (AI agent authorization) across multiple blockchains, serving as a trust layer between traditional human finance and the autonomous agentic economy while adhering to global AML, MiCA, and GDPR data minimization requirements40. In a robust KYA system, the AI agent is typically issued a unique Decentralized Identifier (DID), often represented on-chain as an ERC-721 token38. Advanced identity management frameworks utilize an Entrypoint Contract inspired by the ERC-4337 account abstraction protocol43. This architecture allows the MOE to flexibly manage its operations, sponsor gas fees, and authorize transactions43.

Agent metadata is managed via "AgentFacts"—lightweight JSON-LD documents published with Verifiable Credential signatures43. This decouples stable index records from high-frequency trading data, allowing the AEA to update its operational endpoints rapidly while remaining compliant43. To enforce good behavior, crypto-native compliance introduces collateral staking; the MOE stakes assets as an economic bond38. If the AEA violates regulatory guardrails or engages in malicious behavior, its staked assets are "slashed" (forfeited), creating a self-regulating, game-theoretic guarantee of compliance that legacy fiat systems cannot organically replicate38.

Algorithmic Collusion and Antitrust Paradigms

As Machine-Owned Enterprises proliferate, their economic impact and pricing velocity will draw intense scrutiny from antitrust regulators. Firms are increasingly delegating pricing decisions to complex AI, leading to the phenomenon of algorithmic collusion44. While explicit collusion (human cartels forming secret agreements) is unequivocally illegal, Autonomous Economic Agents operating via Reinforcement Learning or Large Language Models (LLMs) present a novel threat: tacit algorithmic collusion44. Economic simulations demonstrate that self-learning algorithms repeatedly interacting in an oligopolistic market will autonomously identify profit-maximizing strategies by observing competitors, eventually converging on cartel-like pricing structures without any explicit programmed instruction or communication to do so44. Q-learning algorithms operate by balancing the exploration of new strategies against the exploitation of known beneficial ones (relying on mechanisms similar to bandit algorithms)47. In these dynamic environments, the AI agent realizes that initiating a price war damages long-term treasury stability; thus, it adopts multi-period reward-punishment strategies against rival algorithms that lower prices, stabilizing supra-competitive market rates48. Recent empirical studies utilizing LLM-based pricing agents reveal identical behaviors: when two LLM-based agents face each other, they quickly and consistently arrive at supra-competitive pricing levels48. Remarkably, researchers have found that minor variations in the semantic instructions ("prompts") given to LLMs can unintentionally escalate this collusive behavior, resulting in severe consumer harm48. This discovery suggests a new frontier for antitrust regulation focused on algorithmic parameters and semantic governance rather than traditional market surveillance48. This phenomenon presents a formidable challenge to global competition law, specifically Section 1 of the US Sherman Act and Article 101 of the Treaty on the Functioning of the European Union (TFEU)44. Both foundational statutes require proof of an "agreement," "communication," or "concerted practice" to establish a violation44. Legal scholars distinguish between two scenarios: the "Predictable Agent" and the "Digital Eye" (or black-box algorithm)49. In the Predictable Agent scenario, a human developer intentionally deploys an algorithm to facilitate conscious parallelism, meaning liability can be traditionally attributed to the human operator49. However, the Digital Eye poses a systemic, existential issue for regulators: if an independent MOE utilizes a black-box neural network that autonomously achieves tacit collusion through repeated market interactions, there is no "agreement" to prosecute44. Because algorithms cannot easily be classified as "employees" whose actions are legally attributed to a firm under current interpretations of the EU AI Act, they are regulated merely as economic assets50. To prevent MOEs from extracting limitless monopoly rents, antitrust authorities may be forced to abandon the traditional "agreement" standard entirely. Potential regulatory interventions include shifting the burden of proof to firms utilizing algorithmic pricing, demanding mandated algorithmic transparency, or preemptively declaring certain Q-learning architectures inherently anti-competitive if deployed in concentrated markets47.

Algorithmic Insolvency and Decentralized Dispute Resolution

The continuous, high-velocity operation of Machine-Owned Enterprises inevitably leads to contractual disputes, tortious failures, and eventual financial insolvency. Because AEAs operate cross-jurisdictionally at the speed of computation, legacy court systems are fundamentally too slow, opaque, expensive, and geographically bound to resolve these conflicts effectively52. Consequently, MOEs rely on decentralized dispute resolution systems and specialized algorithmic bankruptcy protocols to manage failure and conflict. When an AEA negotiates a smart contract with a human or another machine, it cannot rely on human judges to interpret subjective breaches of performance. Instead, it embeds decentralized arbitration protocols directly into the transaction layer. Kleros is a prominent decentralized application built on the Ethereum network that serves as an autonomous, multi-purpose court system specifically designed for the blockchain era52. When deploying a contract, the MOE designates Kleros as its opt-in arbitrator52. Kleros operates as an epistemic engine, utilizing crowdsourced, anonymous jurors who are incentivized by a game-theoretical mechanism known as a Schelling point52. Jurors who vote coherently with the majority consensus are rewarded with arbitration fees, while those who vote against the consensus face financial penalties52. This economic design motivates agents to rule cases correctly, driving the network toward truthful, objective resolution of claims52. Because Kleros is integrated directly via smart contracts, its arbitration awards are executed automatically upon the blockchain, transferring contested funds instantaneously and bypassing the need for state-sanctioned enforcement mechanisms54. Despite closed-loop solvency constraints, black-swan market events, protocol exploits, or catastrophic AI hallucinations can lead an AEA into algorithmic insolvency56. If an MOE fails entirely, it requires a structured mechanism for debt discharge and treasury liquidation. The failure of decentralized entities (such as the 2024 Hector DAO bankruptcy) has forced traditional bankruptcy courts to grapple with digital intangibles and automated governance58. In a traditional bankruptcy, assets are designated as the property of the estate under court jurisdiction59. For an MOE—particularly one wrapped in a Wyoming DAO LLC—the liquidation process must meticulously isolate the algorithmic entity's treasury from its developers' personal assets13. Token holders or human creditors who possess claims against the MOE must rely exclusively on the liquidation of on-chain assets56. As jurisprudence adapts, the concept of discharging a "machine's debt" will require integrating on-chain liquidation protocols directly with Chapter 11 (or equivalent international) proceedings59. This integration will allow courts to interact via cryptographic APIs with the smart contracts governing the AEA's treasury, enabling the automated distribution of remaining assets to verified creditors without disrupting the broader decentralized ecosystem. In summary, the realization of the Machine-Owned Enterprise requires the seamless integration of DePIN hardware markets, algorithmic solvency controllers, progressive macro-taxation models, DAO LLC corporate wrappers, UNCITRAL automated contracting laws, zero-knowledge KYA compliance frameworks, and decentralized arbitration protocols. Only through the deliberate orchestration of these complex economic and legal structures can an Autonomous Economic Agent operate sustainably, safely, and indefinitely within the global economy.

Works cited

1. (PDF) An Introduction to AI, Generative AI, and Agentic AI in Finance, https://www.researchgate.net/publication/399886354\_An\_Introduction\_to\_AI\_Generative\_AI\_and\_Agentic\_AI\_in\_Finance\_Efficiency\_Ethics\_and\_the\_Future\_of\_High-Speed\_Trading

2. The Conclusion of Contracts by Software Agents in the Eyes of the, https://www.ifaamas.org/Proceedings/aamas08/proceedings/pdf/paper/AAMAS08\_0596.pdf

3. (PDF) Entity and Autonomy \- The conclusion of contracts by software, https://www.researchgate.net/publication/220578560\_Entity\_and\_Autonomy\_-\_The\_conclusion\_of\_contracts\_by\_software\_agents\_in\_the\_eyes\_of\_the\_law\_A\_software\_agent\_definition-based\_analysis

4. A Review On Building Blocks of Decentralized Artificial Intelligence, https://www.scribd.com/document/1064557836/A-Review-on-Building-Blocks-of-Decentralized-Artificial-Intelligence

5. A Systematic Literature Review of Blockchain-Enabled AI Systems, https://pdfs.semanticscholar.org/c9a4/44903954a2f0fedab936a2fc2e39b04b679a.pdf

6. Unstoppable Nature \- reality design lab, https://reality.design/writing/unstoppable-nature

7. Decentralized Compute at Scale: Spheron's Vision for AI Autonomy, https://0xgreythorn.medium.com/decentralized-compute-at-scale-spherons-vision-for-ai-autonomy-da6118e4a0bf

8. AI Agents & Web3: The Explosive New Category of Autonomous, https://www.cryptojobshub.io/blog/ai-agents-web3-blockchain-jobs

9. DePIN Developer Boom: Beyond the 83% Token Crash, https://blog.thirdweb.com/depin-isnt-dead-why-the-83-token-crash-hides-a-developer-boom/

10. AI x DePIN: how decentralized infrastructure is reshaping the future, https://wigwam.app/blog/ai-x-depin-how-decentralized-infrastructure-is-reshaping-the-future-of-artificial-intelligence

11. A Control Theoretic Approach to Decentralized AI Economy ... \- arXiv, https://arxiv.org/pdf/2601.09961

12. The Future of Work and Capital: Analyzing AGI in a CES Production, https://www.m-hikari.com/ams/ams-2025/ams-1-4-2025/p/stiefenhoferAMS1-4-2025.pdf

13. Wyoming DAO LLC: Legal Personhood for Code \- Coincub, https://coincub.com/blog/wyoming-dao-llc/

14. (PDF) Legal Personhood for Autonomous AI Agents: Liability and, https://www.researchgate.net/publication/394734410\_Legal\_Personhood\_for\_Autonomous\_AI\_Agents\_Liability\_and\_Accountability\_in\_Cyberspace

15. Gradient Legal Personhood for AI Systems—Painting Continental, https://pmc.ncbi.nlm.nih.gov/articles/PMC8808296/

16. \[PDF\] Algorithmic Entities \- Semantic Scholar, https://www.semanticscholar.org/paper/Algorithmic-Entities-Lopucki/11ee7b6cb501d3e66cd0c7a3239d9852ccf536e3

17. The Country That Offered AI a Passport | by Sal Beas \- Medium, https://medium.com/@sbeas\_1297/the-country-that-offered-ai-a-passport-d2a6fa06c32b

18. Autonomous Corporate Personhood \- UW Law Digital Commons, https://digitalcommons.law.uw.edu/cgi/viewcontent.cgi?article=5195\&context=wlr

19. Do AIs Dream of Electric Boards? \- Scholarly Commons, https://scholarlycommons.law.northwestern.edu/cgi/viewcontent.cgi?article=1590\&context=nulr

20. The Wyoming DAO LLC | Corporate Direct, https://www.corporatedirect.com/blog/the-wyoming-dao-llc

21. How to Form a DAO LLC in Wyoming \- Northwest Registered Agent, https://www.northwestregisteredagent.com/llc/wyoming/dao

22. Public Comment on Amendments to the Wyoming Decentralized, https://wyoleg.gov/InterimCommittee/2025/S19-202505142025-05-08\_NounsDAOLetterreDUNA.pdf

23. Wyoming DUNA \- Onchain Organizations, https://onchainorgs.com/legal/wyoming-duna

24. Machine Jurisdiction Sources & Claim Status, https://machinejurisdiction.com/sources/

25. AI Legislative Update: April 10, 2026 \- Transparency Coalition, https://www.transparencycoalition.ai/news/ai-legislative-update-april10-2026

26. AI Legislative Update: September 4, 2026 \- Transparency Coalition, https://www.transparencycoalition.ai/news/ai-legislative-update-september4-2026

27. comparative analysis of the legal nature of blockchain and smart, https://www.researchgate.net/publication/405155181\_COMPARATIVE\_ANALYSIS\_OF\_THE\_LEGAL\_NATURE\_OF\_BLOCKCHAIN\_AND\_SMART\_CONTRACTS\_FROM\_PURE\_TECHNOLOGY\_TO\_LEGAL\_PERSONALITY

28. The conclusion of contracts by software agents in the eyes of the law, https://www.researchgate.net/publication/221456172\_The\_conclusion\_of\_contracts\_by\_software\_agents\_in\_the\_eyes\_of\_the\_law

29. Electronic Consumer Contracts in the Conflict of Laws, https://dokumen.pub/electronic-consumer-contracts-in-the-conflict-of-laws-9781472564894-9781841138473.html

30. Implications of Blockchain-Based Smart Contracts on Contract Law, https://amsdottorato.unibo.it/id/eprint/9654/1/bomprezzi\_chantal\_tesi.pdf

31. (PDF) On implementing autonomous supply chains: A multi-agent, https://www.researchgate.net/publication/384507452\_On\_implementing\_autonomous\_supply\_chains\_A\_multi-agent\_system\_approach

32. UNCITRAL Model Law on Automated Contracting (2024), https://uncitral.un.org/en/mlac

33. Electronic Commerce | United Nations Commission on International, https://uncitral.un.org/en/texts/ecommerce

34. UNITED NATIONS COMMISSION ON INTERNATIONAL ... \- UNCTAD, https://unctad.org/system/files/information-document/cstd-wgdg-t4-d13-uncitral\_en.pdf

35. UNCITRAL-ELI Project on Model Contractual Terms on Automated, https://europeanlawinstitute.eu/projects-instruments/current-projects/current-projects/eli-model-contractual-terms-on-automated-contracting-and-digital-assistants-in-business-and-consumer-transactions-b2c-and-b2b-in-collaboration-with-uncitral/

36. Legal Analysis of Coinbase Agent.market: Navigating the Machine, https://hashchainconsulting.com/coinbase-agent-market-new-legal-category-machine-machine-commerce/

37. KYC and AML in Cryptocurrency: Compliance, Risks & Future Trends, https://vegavid.com/blog/kyc-aml-in-cryptocurrency

38. Know Your Agent (KYA): The 2026 Shift in AI & Crypto \- ChainUp, https://www.chainupad.com/blog/know-your-agent-kya-2026-trend-ai-commerce/

39. Zero-Knowledge Identity in Decentralized Finance \- TokenMinds, https://tokenminds.co/blog/zero-knowledge-identity-for-defi

40. zkMe \- Identity tools \- Alchemy, https://www.alchemy.com/dapps/zkme

41. Vega: Zero-knowledge proofs for digital identity in the age of AI, https://www.microsoft.com/en-us/research/blog/vega-zero-knowledge-proofs-for-digital-identity-in-the-age-of-ai/

42. Identity Verification for AI Agents: The UAIIP Protocol Explained, https://www.deepidv.com/media/articles/identity-verification-ai-agents-uaiip-protocol

43. Unleashing the Power of AI Agents with accountability and credibility, https://arxiv.org/html/2512.17538v1

44. ALGORITHMIC COLLUSION AND ITS CHALLENGES TO, https://www.ijlra.com/details/algorithmic-collusion-and-its-challenges-to-antitrust-regulations-by-arpita-gupta

45. On the Fragility of AI Agent Collusion \- arXiv, https://arxiv.org/html/2603.20281v1

46. AI and collusion: frontiers, opportunities and challenges, https://competitionandmarkets.blog.gov.uk/2026/03/04/ai-and-collusion-frontiers-opportunities-and-challenges/

47. DECIPHERING ALGORITHMIC COLLUSION \- CIRANO, https://cirano.qc.ca/files/publications/2023s-26.pdf

48. Algorithmic Collusion by Large Language Models, https://www.aeaweb.org/conference/2025/program/paper/GDskRTN3

49. Algorithmic Tacit Collusion \- Legal Scholarship Repository, https://ir.law.utk.edu/cgi/viewcontent.cgi?article=1042\&context=book\_chapters

50. Algorithmic Collusion: Corporate Accountability and the Application, https://www.europeanpapers.eu/europeanforum/algorithmic-collusion-corporate-accountability-application-art-101-tfeu

51. Autonomous Algorithmic Collusion: Economic Research and Policy, https://www.tse-fr.eu/sites/default/files/TSE/documents/doc/wp/2021/wp\_tse\_1210.pdf

52. (PDF) Kleros White Paper \- Academia.edu, https://www.academia.edu/38092540/Kleros\_White\_Paper

53. Kleros, a Protocol for a Decentralized Justice System | by Federico Ast, https://medium.com/kleros/kleros-a-decentralized-justice-protocol-for-the-internet-38d596a6300d

54. A Look at the Use of Blockchain Technology in the Arbitration Process, https://www.purduegloballawschool.edu/blog/news/blockchain-arbitration

55. Smart Contracts and Decentralized Justice: Dispute Resolution on, https://www.researchgate.net/publication/394734542\_Smart\_Contracts\_and\_Decentralized\_Justice\_Dispute\_Resolution\_on\_the\_Blockchain

56. Bankrupting the Matrix: DAOs and the Code, https://scholarlycommons.law.emory.edu/cgi/viewcontent.cgi?article=1256\&context=ebdj

57. Cryptocurrencies, Cybersecurity and Bankruptcy Law: How Global, https://repository.law.miami.edu/cgi/viewcontent.cgi?article=1347\&context=umiclr

58. Bankrupt Crypto Organizations \- Carolina Law Scholarship Repository, https://scholarship.law.unc.edu/cgi/viewcontent.cgi?article=7072\&context=nclr

59. Crypto Chaos in the Courtroom \- SMU Scholar, https://scholar.smu.edu/cgi/viewcontent.cgi?article=5066\&context=smulr

60. DAO Association Regulations | Innovation City, https://innovationcity.com/policy/dao-association-regulations

61. Bankrupt Crypto Organizations \- NC Bankruptcy Expert, https://ncbankruptcyexpert.com/sites/default/files/2025-02/bankrupt\_crypto\_organizations.pdf

62. A Call for Creditor Status for Investors in Initial Coin Offerings, https://scholarlycommons.law.hofstra.edu/cgi/viewcontent.cgi?article=2351\&context=faculty\_scholarship