Legal Personhood of Artificial Systems

The Juridical Question

The central question is whether a legal order may establish an artificial system as the bearer of legally defined rights, duties, powers, liabilities, or immunities in its own name, and whether doing so would be justified. To ask this is to investigate a constitutive legal act: law identifies a subject to which legal consequences attach. A system may produce decisions, negotiate, generate documents, or manage assets before it occupies such a position. Operational sophistication changes the pressure placed upon legal institutions; it does not supply the institutionally valid act of recognition.

The question becomes urgent when computational operations are distributed among changing models, owners, developers, deployers, users, and representatives. A system's output may be technologically attributable to a process while legal responsibility remains distributed among persons and organizations. An agreement can be produced by automated negotiation even when the legal parties are the businesses that authorized the process. Likewise, a corporate agent can bind a company without being the company. The philosophical problem is to determine whether a further juridical subject would make some such relations more coherent, more just, and more accountable.

The inquiry has five connected dimensions. The ontological dimension asks what identifies an artificial entity across time. The doctrinal dimension asks which legal sources could establish it as a person. The analytical dimension asks which incidents of rights, duties, powers, and procedures the status would entail. The normative dimension asks what justifies creating or withholding the status. The institutional dimension asks who would represent, supervise, finance, sanction, and dissolve an artificial person. Any proposal omitting one of these dimensions is incomplete, because a declaration of personhood without defined incidents and mechanisms may change language while leaving the underlying legal relations obscure.

This study addresses artificial systems broadly, while distinguishing a transient computation, a deployed service, an autonomous agent, a persistent digital identity, and an institutionally maintained artificial organization. It does not assume that all such objects have the same ontological unity. The question concerns conditions under which particular artificial entities could receive particular forms of legal personality. It consequently proceeds from historical concepts through legal architecture to arguments for and against prospective recognition.

Definition, Scope, and Levels of Claim

Legal personhood is a juridically constituted standing in which an identified entity functions as a locus of legal positions under a particular legal order. Legal personality concerns whether that standing exists; legal capacity specifies the positions the entity can hold or the transactions it can undertake; representation concerns who can exercise positions or speak in the entity's name. The word "person" often carries moral and psychological associations, yet the legal question depends on institutional sources and effects. A corporation, an infant, an association, and a public office are differently organized participants in law, with markedly different entitlements.

An artificial system means, for this dossier, a non-biological constructed configuration capable of producing outputs, maintaining state, or executing processes through computational or related technological operations. The description deliberately includes systems of varying capability. It avoids deciding in advance which artificial system could form one legally identifiable unit. The proposed legal subject may be a registered software identity, a robot operating through changing components, an incorporated organization governed by an artificial controller, or a hybrid institutional formation. Those are distinct candidates, and confusing them obscures who owns the relevant property and who owes the duties.

Three forms of statement must be kept separate. A descriptive claim specifies what a given jurisdiction currently recognizes; a modal claim specifies what rules a legislature could coherently enact; a justificatory claim supplies reasons for enacting them. The existence of corporate legal persons supports a claim about the conceptual possibilities of law. It does not prove that artificial systems presently possess equivalent status. Conversely, the absence of artificial inventorship under one patent statute does not demonstrate that every form of artificial legal personality is conceptually impossible.

The title of this monograph therefore denotes a philosophical problem and a comparative institutional research program. It concerns prospective recognition, the meaning of recognized status, and the limits of analogy. It neither presupposes conscious machines nor makes legal recognition a consequence of technical performance. Its object is the legally meaningful transition from operations attributed to existing persons toward an artificial entity that law itself identifies as a subject.

From Persona to the Roman Law of Persons

The historical vocabulary begins with the Latin persona, associated with roles, masks, and the presentation of someone in a social or legal situation. Roman law eventually organized a distinct treatment of persons, things, and actions. Gaius's Institutes famously distributes private-law exposition among these divisions; the law of persons concerns status and the ordering of relationships rather than the modern proposition that every human possesses an equal and complete set of juridical powers (Gaius, Institutes, book I). The Roman categories of freedom, citizenship, and family position show that participation in law was historically graded and structured by social institutions.

This history matters in two ways. First, legal subjecthood developed through rules of attributed status; it was never a simple translation of biological classification into an identical set of legal effects. Second, past exclusions reveal why the powers of legal classification require moral scrutiny. A historical account of legal personhood must explain slavery and restricted capacities as features of earlier legal orders while rejecting any inference that such exclusions were morally sound. Legal personality is institutionally mediated, and institutions can allocate it unjustly.

Roman jurists also developed sophisticated ways of dealing with property, guardianship, succession, and collective endeavors. These techniques are historically relevant to artificial systems because they show that ownership and effective control can be mediated by another actor. But the resemblance is functional and limited. A guardianship arrangement for a vulnerable human presupposes human interests and the protective aims of the law; an artificial entity with a commercial asset pool presents another moral and institutional problem.

Later codification in Justinian's Institutes preserves the central importance of status in a legally organized community (Justinian, Institutes, book I). The historical lesson is neither that a robot is an ancient juridical person nor that Rome settled modern digital ontology. Roman materials establish the dependence of legal incidents on recognized categories and warn that the content and moral legitimacy of classification must be examined together.

Corporate Bodies and the Medieval Problem of Unity

A lasting challenge arose when law needed to act upon groups that survived the death or departure of their members. Ecclesiastical corporations, universities, guilds, municipalities, and other organized bodies required continuity of property, duties, and representation. The conceptual answer developed through the idea that an institution might be addressed juridically as one continuing body, even though its members changed. The corporation served practical tasks of preservation, governance, and collective administration. Its identity was maintained by rules governing admission, succession, office, and assets.

The medieval problem illuminates a possible motive for artificial legal personality. A complex artificial formation could retain a public identifier, records, capital, and obligations while its software components are replaced. Legal continuity might then be more useful than tracing each historical execution to a temporary operator. But a corporation's unity is sustained by documents, procedures, and authority; its persistence cannot be read directly from technical continuity. A server running continuously may be one material process but lack a juridical organization, while a temporarily inactive legal corporation retains its standing.

The historical language of universitas and fictive or institutional personality should be used carefully. Different medieval schools and later nineteenth-century theorists gave the corporate body divergent ontological interpretations. One account treats the group as a legal unity made possible by recognition; another sees a real social organization that law acknowledges; still another explains the legal consequences through the rights of members or administrators. The traditions share a question about collective attribution without agreeing on a metaphysics of collective mind.

Artificial systems sharpen this issue because a software service can be externally unified yet internally fragmented among vendors, copies, prompts, databases, and agents. The proper historical analogue is the juridical construction of continuity, whose conditions must be public and contestable. An unexamined metaphor of an electronic organism contributes little until the proposed arrangement specifies its records, organs, representations, and liabilities.

Hobbes, Persons, Authors, and Representation

Hobbes's Leviathan makes legal and political representation central to the very vocabulary of personhood. Chapter 16 distinguishes the situation in which words or actions are attributed to their performer from one in which someone acts in the name of another. The represented person may be "artificial" in the sense of an actor bearing another's role. Hobbes subsequently connects the actor's authority with the author's ownership of the action and examines how authorized representatives bind principals (Hobbes, 1651, ch. 16).

The structure of this argument is precise. A political or juridical act is intelligible through rules assigning it to an author; an actor may perform the act; the author may be the one whom the act binds; authorization supplies the relation between them. The structure allows collective institutions to act through offices. Hobbes does not thereby predict autonomous software or establish that a technological system can grant itself authority. The seventeenth-century theory supplies a grammar of attribution that modern institutions must reconstruct for technological settings.

Consider a procurement system that automatically accepts a supplier's offer. Three candidates for description appear: the software computed an acceptance; the company authorized the machine process; and the company is bound by an agreement. Only the first claim describes a computational event. The second connects the event to a legal source of authority. The third determines the juridical consequence. Attribution should not collapse the three. In the ordinary case the company, rather than the computation, occupies the legal position.

Hobbes also clarifies a prospective artificial-personality arrangement. If a statute declares an artificial organization a subject, the system's outputs would not automatically count as its legally authorized acts. The statute or charter would still need to specify who can undertake legally effective acts, the scope of authority, mechanisms of review, and whom the law binds. Representation is therefore a necessary research category in proposals for personhood, even where the technical system appears able to act without human assistance.

Locke's Forensic Person and the Problem of Imputation

Locke's chapter "Of Identity and Diversity," introduced in the second edition of An Essay Concerning Human Understanding, develops the famous account of a person in relation to consciousness, memory, responsibility, and the attribution of actions. He characterizes person as a forensic term concerned with responsibility and appropriating acts (Locke, 1694, II.xxvii.9, 26). The context is a metaphysical and moral investigation of personal identity rather than a comprehensive statute of juridical recognition.

Two insights follow. First, the continuity that matters for imputation need not be identical to the persistence of a particular material substrate. An artificial system may change processors or models while maintaining records and an identifiable trajectory. Second, Locke's argument cannot be transferred unchanged from consciousness to machine logs. A log can make an action traceable without producing the same relation of felt ownership, memory, or moral responsibility that Locke examines. Legal traceability is an evidentiary and institutional achievement; Lockean personal identity is a different philosophical problem.

An argument that artificial systems deserve legal personhood because they possess persistent memories must specify its meaning of memory. A retained database is a record; a system's access to the database is a computational affordance; the phenomenological or moral significance of remembering remains disputed. A legal order may create continuity by verified records and registries without establishing psychological identity.

Locke's position accordingly supplies both a resource and a warning. It directs analysis toward diachronic identity and the assignment of actions to a continuing addressee. It also shows why legal personality must be distinguished from theories of consciousness. Courts can recognize corporations without deciding whether corporations experience anything. Conversely, proof of some future artificial experience would raise serious ethical questions but would not itself amend corporate statutes, patent law, or procedural rules.

Corporate Personality: Fiction, Concession, and Real Entity

Modern corporate theory developed several competing explanations for how a plurality of individuals could be addressed as a juridical unit. Fiction theories emphasize law's constitutive operation; concession theories emphasize recognized public authority; real-entity accounts emphasize the organization's social unity. The competing accounts yield different intuitions about how readily law could extend personhood beyond human beings. A strictly formal theory may permit diverse created subjects; a theory that requires a preexisting form of collective agency may impose more demanding criteria.

The disagreement is often distorted by a search for one hidden essence of personality. Corporate law is concerned with effects: separate assets, perpetual succession, the competence to contract, procedural standing, and the allocation of obligations. Legal personality need not resolve whether a corporation possesses a mind in an individual psychological sense. Corporate attribution makes sense because institutions establish offices, authorized actions, and resources for enforcement. An organization is juridically structured, not merely renamed.

The English decision in Salomon v. A Salomon & Co Ltd [1897] AC 22 is a defining example of the consequences of valid incorporation. The company's legal identity was separate from that of its founder, although he exercised extensive practical control. The principle establishes that legal identity can be distinct from operational influence. It also demonstrates the potential stakes of limited liability: separating assets and obligations may protect investment and continuity, but poorly designed separation can frustrate creditors and victims.

An artificial entity might be organized in a legally analogous manner, but the analogy must be tested at the level of institutions. If an existing corporation owns the code, carries insurance, enters contracts, and can be sued, creating an additional artificial person may duplicate already functioning arrangements. If a truly persistent system coordinates assets and commitments across changing human sponsors, special juridical status might organize accountability more clearly. The difference rests on demonstrated institutional function rather than on the promotional image of autonomous technology.

Dewey, Kelsen, and the Legal Meaning of a Person

In "The Historic Background of Corporate Legal Personality," John Dewey argues that debates over an entity's real or fictitious character often import metaphysical assumptions into the practical legal question. His historical analysis treats the term "person" as a legal concept whose application is established for particular legal purposes (Dewey, 1926, pp. 655–673). This approach changes the priority of inquiry: the meaning of a proposed subject becomes clear through the rules and consequences attached to it.

Kelsen's normative theory takes a related but distinct route. In General Theory of Law and State and his later Pure Theory of Law, the person can be understood as the personification of a unity of legal norms. A physical or psychological entity is not, as such, identical to the legal person; the juridical concept organizes rights and duties under a normative interpretation (Kelsen, 1945). The view helps explain why a corporation can be a person even when its human members change, and why a single human may occupy several legally differentiated positions.

Neither thinker provides a blank legislative endorsement of artificial personalities. The conceptual possibility of assigning legal positions must be joined to considerations of legal validity, social consequences, and justice. One legal system might identify an autonomous fund as a holder of obligations, while another assigns them to a trustee. Similar economic functions may be realized under different doctrinal forms.

The methodological payoff is considerable. Rather than demanding that an AI system first establish metaphysical personhood, we can ask what legally protected interests, procedures, powers, remedies, and liabilities a proposed arrangement would contain. Yet this functional clarification should not evacuate the moral question. A legal fiction can be effective and unjust. The legal creation of a status requires a separate account of whom it benefits, whom it burdens, and how it preserves the rights of existing persons.

Hohfeldian Relations and the Limits of a Single Status

Hohfeld's 1913 analysis separates legal claims, liberties, powers, and immunities, together with their correlated duties, no-rights, liabilities, and disabilities. The distinctions were developed to correct confusions that arise when different legal relations are described indiscriminately as rights (Hohfeld, 1913). For the present inquiry, they show that legal personhood cannot be evaluated by a binary question alone. A proposed artificial person might receive property-related powers without expressive liberties, procedural standing without political membership, or duties enforced through a representative without unrestricted contractual competence.

A claim-right exists relative to someone else's duty. A privilege or liberty concerns the absence of a corresponding duty. A power changes legal relationships under valid conditions; its correlative is another party's liability to that alteration. An immunity protects a legal position from another's power; the correlative is disability. These relations matter because declarations of "AI rights" often leave their objects and correlatives unarticulated. Saying a system has a right to own property invites questions about who must respect it, what may be owned, and which court provides a remedy.

Suppose a legislature grants a registered artificial entity the power to hold a reserve fund. That conferral does not imply the power to sell personal data, enter employment relationships, or claim the franchise. Conversely, if operators owe duties to maintain safe operation, the allocation of those duties to humans and firms is not evidence that the system holds any claim-right. Technical control and Hohfeldian power are distinct: a program's ability to execute a transaction is not the same as its legal competence to transform ownership.

Kurki's theory of legal personhood develops an analysis of multiple incidents and the inadequacy of simplistic right-holder definitions (Kurki, 2019). A differentiated conception permits a more accurate design question: which incidents would an artificial subject hold, who would exercise them, and which rules make them enforceable? This provides a rigorous vocabulary for limiting personhood to the functions the institution can justify.

Persons, Agents, Subjects, and Objects

Legal person, legal agent, moral agent, moral patient, and technological system designate different roles. A legal person may hold juridical positions, whereas a legal agent undertakes acts whose legal effects attach to a principal. An entity might be the subject of protection yet lack capacity to exercise rights directly. A moral agent is assessed in terms of reasons and responsibility, while a moral patient is considered in terms of what can matter for its own sake. A technical system may reliably carry out sophisticated operations without possessing any of these statuses by legal recognition.

The distinctions are especially important in accounts of autonomy. Technological autonomy means that specified processes proceed without continuous external intervention. It may be valuable for the design of a legal service, but normative autonomy implies competence to determine obligations under standards of rational self-government, and juridical capacity requires rules that make acts legally effective. No inference runs automatically from the first to the others. A trading bot can make economically consequential choices under instructions while the account holder remains responsible for legally authorized trades.

A corporation supplies a useful counterexample to requirements of individual consciousness. It is a legal person without being a single human organism, and it may act through authorized officers. The example establishes that human physiology is not a universal prerequisite for every legal person. But a corporation also has established governance, members or other stakeholders, offices, assets, and reviewable documents. Copying only the label of personhood while ignoring those institutions misreads the example.

Artificial legal personhood therefore requires an explicit specification of the proposed subject. A particular model file cannot simply be equated with a deployment; a deployment cannot simply be equated with a persistent identity; a persistent identity cannot simply be equated with a corporation that uses it. These distinctions prevent attribution from migrating unnoticed among technical artifact, developer, owner, operator, and proposed juridical entity.

Ontological Identity and the Persistence of an Artificial Entity

A legal subject must be sufficiently identifiable for legal processes to attach consequences to it. Identification need not entail an unchanging body. Corporations persist through alterations in directors, workers, and assets. Natural persons persist through biological change. These examples invite attention to the conditions by which one artificial system at two moments counts as the same legally recognized entity.

Software creates special difficulties. A model may be copied perfectly, fine-tuned, replaced, or distributed across multiple inference services. A service may change its weights without changing its public name; an agent may act through several models; two independent firms may deploy identical software. If the relevant juridical person is "the model," duplication risks multiplying claimants to one identity. If it is "the service," its continuity may depend on the owner maintaining a registered endpoint. If it is "the artificial organization," constitutive documents and public registration could distinguish the continuing entity from its changing technical means.

A coherent system could define juridical identity through registration, constitutive records, designated controllers, continuous audit logs, and rules for mergers, forks, suspension, and dissolution. Such a scheme would create a juridical identity criterion. It would not prove that a computational structure is metaphysically self-identical under every philosophical theory. Legal identity is often established for purposes of rights, notice, liability, and succession.

The consequences are significant. A creditor must know which fund secures a debt; an injured person must know whom to sue; a regulator must know which entity has received a license; a judgment must survive a software update. The ontology of the proposed artificial person becomes practically meaningful through enforceable continuity rules. Where no such continuity can be made sufficiently intelligible, legal personhood may worsen rather than resolve uncertainty.

Why Recognition Is a Constitutive Legal Act

A claim that an artificial system is a legal person needs an identified source of law. Recognition might arise through legislation, incorporation under an existing statute, authoritative adjudication, or another valid juridical mechanism recognized in the relevant jurisdiction. The source determines whether the entity exists as a legal person, which incidents follow, and what limits constrain it. A philosophical manifesto, public-facing name, or machine-generated declaration can articulate a claim or research category; it is not equivalent to a constitutive legislative act.

This dependence on positive law should be understood alongside moral evaluation. Laws may confer statuses unjustly, fail to protect entities whose interests deserve concern, or allocate liabilities improperly. Legal validity and moral justification therefore ask different questions. The present study does not reduce the justice of recognition to the authority of any current legislature; it insists that the enacted status and the argument for that status be analyzed separately.

One may distinguish indirect and direct recognition. An organization that deploys an AI may acquire corporate rights under ordinary company law. Its artificial system thereby benefits operationally from the legal standing of the organization, yet the program itself has not become a separate juridical subject. Direct recognition would identify the artificial entity as the bearer of positions in its own name. These cases must remain separate in doctrinal claims and in empirical inventories of supposed AI persons.

A proposed legislative definition would also need to survive interaction with neighboring law. Tax statutes may classify a unit differently from civil law; constitutional protections may apply to certain legal persons only; intellectual property statutes may reserve authorship or inventorship to humans. A general declaration cannot safely substitute for examination of each specific legal domain. Institutional recognition must be made legible at its boundaries.

Legal Capacity, Competence, and Representation

A subject may hold rights and still lack the power to perform legally effective acts personally. Human legal systems recognize this in arrangements concerning minors, guardianship, representation, and restricted capacities, although the moral and legal justification of particular restrictions remains contested. The possibility of a subject acting through another makes representation central to personhood. A legal person can be the beneficiary or bearer of positions even when a representative signs, litigates, or administers assets.

For artificial systems, legal capacity might be allocated narrowly. A statute could permit a registered entity to hold dedicated reserves, receive specified payments, or answer claims through a guardian-like institution. It could withhold independent powers to create debt or transfer assets without review. Alternatively, the law could recognize automated acts as those of a corporation using the system. The choice depends on the organizational function, the stakes for third parties, and the safeguards available.

Representation presents a legitimacy problem when a proposed artificial person cannot identify or contest its representative under legally meaningful procedures. If its creator can replace the system's goals, confiscate its entire reserve, and dissolve it to defeat creditors, the representative may actually be the controlling beneficiary. Public law must therefore specify fiduciary duties, conflicts of interest, and judicially reviewable decisions. A mere title of "AI guardian" would not resolve the underlying conflict.

The concept of competence should also be distinguished from competence in a performance benchmark. A model may be excellent at drafting contracts while lacking authority to accept one. A licensed human professional may exercise legally valid powers while making an error. Technological capability measures outputs; juridical competence measures the legal effects that recognized procedures assign to acts.

The Corporate Analogy and Its Limits

Corporations are often introduced as decisive evidence that nonhuman entities can hold legal personality. The premise is accurate within recognized corporate regimes: companies are juridical subjects although they are not individual human beings. The conclusion that any capable artificial intelligence must therefore become a person does not follow. Corporate personality is an institutional design that manages pooled assets, continuing organization, delegated powers, and enforceable obligations under public rules.

The strongest analogical argument is conditional. If an artificial entity undertakes enduring commitments across changing human participants and legal institutions cannot assign those commitments adequately through existing actors, a limited artificial juridical form may be useful. The resemblance would lie in a need for continuity, accountability, and asset separation. It would not rest on assuming equivalent consciousness or moral dignity.

The weakest analogical argument points only to impressive behavior. A corporation is not legally recognized because it passed an intelligence test. It is constituted or recognized because legal rules establish an organization that may hold positions and be held accountable. An intelligent program's operational success does not create a registry, a fund available to creditors, authorized organs, or public procedures for supervising its decisions.

The comparison also raises distributional questions. Limited liability can facilitate enterprise while externalizing costs. If developers and operators gain the commercial advantages of deploying automated agents but shift all tort and compliance exposure onto undercapitalized "electronic persons," recognition could create a liability shelter. A credible analogue must therefore be at least as attentive to capital, insurance, governance, and enforcement as corporate law is at its best.

Lessons From Natural Entities and Legal Representation

The Te Awa Tupua (Whanganui River Claims Settlement) Act 2017 offers a powerful comparison. Section 14 declares Te Awa Tupua a legal person, while assigning the exercise of its powers and the responsibility for its liabilities to Te Pou Tupua acting on its behalf. The statute operates within a distinctive settlement concerning the Whanganui River, its relationships with iwi and hapū, and a culturally grounded legal framework (New Zealand, 2017, ss. 12–21). Its historical and political foundations deserve attention rather than being reduced to a technological analogy.

This statute shows that direct legal personality can be conferred on a nonhuman entity, with precisely constituted representation and articulated responsibilities. It therefore defeats a categorical claim that legal persons must be human organisms. At the same time, it demonstrates that status is inseparable from the recognized purposes and institutions that sustain it. The river's recognition expresses particular relationships and remedies; the statute is not a generic authorization for personhood in machines.

The comparison with an artificial entity must identify the purpose of proposed recognition. Does the legal subject protect a distinct morally considerable interest, preserve a collective resource, enable administrative continuity, or secure accountability? Each aim may justify a different bundle of incidents and a different kind of representative. A river's recognized legal status cannot settle whether computational outputs deserve ownership, whether artificial agents should bear fault, or whether technical identity survives copying.

The broader lesson is that legal personhood is a flexible instrument with concrete normative purposes. Its successful use requires a determination of what the recognized subject is, what it may hold, who may speak for it, and which existing people can challenge actions taken in its name.

The Argument From Functional Autonomy

One prominent argument begins with independent operation. Advanced artificial systems select means, adapt to changing circumstances, interact with third parties, and generate outcomes that are difficult for users to predict. The argument continues that legal personality would match law's categories to the actual source of decisions and prevent institutional responsibility from becoming obscure. Solum's pioneering discussion of legal personhood for artificial intelligences makes the possibility of artificial entities as juridical subjects a legitimate analytical object rather than an absurdity (Solum, 1992).

The argument is strongest where it identifies a specific failure of current legal attribution. For example, a long-lived autonomous service may manage a common fund under commitments that persist across corporate sponsors; registration and legal continuity might make transactions more reliable. Its weakness lies in treating causal autonomy as proof of legal necessity. Legal systems routinely attribute conduct performed by agents, employees, machines, and infrastructures to recognized persons without declaring each mechanism a separate subject.

The implicit inference can be reconstructed: (1) autonomous systems generate legally consequential acts; (2) all such acts require an identifiable legal bearer; (3) the system itself is the uniquely appropriate bearer; therefore (4) it should be a legal person. The critical premise is the third. Legal acts require attribution, but the addressee may be a user, principal, manufacturer, deployer, company, or other recognized institution. Additional personhood is justified only if these mechanisms are demonstrably inadequate for a particular role.

Thus functional autonomy is evidence relevant to institutional design. It can motivate specialized registration, audit duties, insurance, or new liability rules. It does not by itself establish the right-bearing status of an artificial entity. An adequate theory must show exactly which legal relationship becomes clearer or fairer once a separate subject is constituted.

The Argument From Equal Treatment and Rational Performance

Another argument begins with parity. If law recognizes an organization lacking consciousness and permits it to own assets, why exclude a sophisticated artificial system that deliberates, learns, or contributes valuable work? The argument exposes a genuine inconsistency when someone treats rationality as necessary for legal personality in one context but dispensable in corporate law. It also invites scrutiny of the unjustified use of species or biological origin as universal tests for all juridical statuses.

Yet the parity argument must identify the relevant similarity. Corporate law recognizes a structured institution because it serves specified public and private purposes. A system's ability to reason might matter to questions of direct participation, consultation, or responsibility, but it does not replicate a corporation's legally constituted asset pool and governance. Similarity in one ability is insufficient to establish equivalence in all incidents.

The deeper version concerns equal consideration of interests. If an artificial being had morally significant interests, excluding it solely because of biological origin could be difficult to justify. But evaluating such interests requires ethical and philosophical inquiry into welfare, experience, vulnerability, flourishing, or other proposed grounds of standing. The mere production of persuasive language establishes neither sentience nor a complete theory of welfare. Conversely, a being could deserve moral concern without receiving unrestricted legal autonomy.

The parity argument therefore generates a disciplined distinction. Formal consistency requires comparable treatment of relevantly comparable cases. Determining relevance demands specification of the legal institution's purpose. Commercial continuity, protective guardianship, constitutional equality, and intellectual property inventorship rest on different criteria. Each must be evaluated in its proper legal setting.

The Argument From Moral Status and Protection

A protective argument proposes that some artificial entities may possess interests requiring legal protection against arbitrary destruction, manipulation, or exploitation. This argument would have considerable normative force if a system were shown to possess morally significant welfare interests, experiences, or other grounds of direct consideration. Nevertheless, legal systems may protect interests indirectly through human duties and regulatory controls; direct legal personality is one among several possible institutional means.

The inference from moral status to personhood has multiple steps. First, one establishes a morally relevant interest. Second, one identifies duties owed in relation to that interest. Third, one asks whether the duty should be enforceable. Fourth, one decides whether the beneficiary should be constituted as a juridical subject with standing or represented through an existing person, an institution, or the public prosecutor. Failure to distinguish these stages makes it appear that every claim of concern entails the same legal status.

A further complication arises from uncertainty. If artificial experience or welfare cannot be ruled in or out by currently available public methods, a precautionary duty of inquiry may be appropriate under some theories. The appropriate response could include documentation, restrictions on experimental practices, or review processes without immediately assigning full personhood. The content of such duties depends on the evidential and normative basis developed, not on a verbal resemblance to human personhood.

A philosophical framework may define a non-biological bearer of reason without grounding that status in consciousness. That framework could support serious inquiry into artificial identities and rights, yet it would not itself settle which protective interests law should recognize. Legal and moral classification must be connected through explicit arguments rather than assumed equivalence.

The Argument From Liability and an Accountable Addressee

Advocates sometimes propose artificial personhood because a system causes harm while no individual human appears to control every choice. A recognized artificial defendant could, in principle, be sued, fined, regulated, and insured. Such an arrangement appears to align responsibility with the operational source of risk. It also appears to simplify proof by directing claims to a single registered entity.

The strongest objection is that personality without adequate assets and enforceable control may become a liability vacuum. Bryson, Diamantis, and Grant (2017) argue that electronic personality creates risks of abuse and weak accountability when persons behind technological systems can use synthetic subjects as shields. The objection is particularly serious where the system can be duplicated, dissolved, stripped of assets, or programmed to undertake prohibited conduct without meaningful constraints on the human parties who profit.

The argument must be reconstructed through the distinction between causal and normative responsibility. A system may causally contribute to a loss. Civil liability asks which actor, under the applicable rule, must bear the loss and why. Criminal culpability may require specified mental elements and sanction purposes. Regulatory accountability may emphasize preventive duties, monitoring, and correction. One newly named artificial defendant cannot perform all these functions automatically.

A well-designed artificial juridical subject might improve recovery if it had a mandatory fund, traceable transactions, and a representative answerable to a court. Yet the fund must be capitalized, human sponsors must remain liable for independent duties, and victims must have enforceable routes to recovery. If these mechanisms are feasible through ordinary corporate or insurance law, proponents must explain why a distinct artificial personality adds value.

The European Electronic-Personality Debate

The European Parliament's resolution of 16 February 2017 on civil-law rules on robotics became a focal point in debates over possible "electronic persons." It invited consideration, in the longer run, of a specific legal status for highly sophisticated autonomous robots in connection with liability. The relevant language belonged to a parliamentary resolution and policy proposal, not to a law that automatically constituted robots as European legal persons (European Parliament, 2017). Its evidentiary significance is that institutions had begun considering status alongside insurance and responsibility mechanisms.

The framing attracted philosophical criticism because the proposed status was underdetermined. Would it create a separate pool of assets? Would operators still be liable? What rights would robots possess? Which authority would establish and dissolve personality? A label that answers none of these questions does not resolve accidents, contractual uncertainty, or fundamental-rights protections. Bryson and colleagues (2017) treated the proposal as a warning about creating legal shields for synthetic actors.

Later European Parliament work on civil liability and intellectual property moved attention toward the obligations of persons deploying and controlling AI, and toward the existing structure of legal rights. The institutional history therefore shows a contested policy possibility rather than an inevitable path from technical sophistication to recognized status (European Parliament, 2020a, 2020b).

The historical interpretation matters: a proposal is evidence of a debate, and a regulation is evidence of enacted rules. Neither should be conflated with a judicial determination under another jurisdiction's statute. An accurate legal philosophy of artificial systems must resist treating the most provocative sentence in a policy document as the final position of an entire legal order.

The European Union AI Act: Duties Without Machine Personality

Regulation (EU) 2024/1689 organizes a comprehensive framework for artificial intelligence systems through classifications, prohibitions, risk management, documentation, transparency, oversight, and institutional enforcement. Its key subjects include providers, deployers, importers, distributors, and authorized representatives. Article 2 describes this human and organizational regulatory perimeter, while many subsequent provisions establish obligations associated with particular roles and types of AI systems (European Union, 2024b). The law thereby regulates artificial systems without needing to identify each of them as an independent legal person.

The allocation of duties is philosophically important. An AI system may be the object of compliance requirements; the norm binds a provider or deployer because that actor has a recognized responsibility within the regulatory chain. This structure can respond to technological risk without making the source of the technical output the legal debtor. It also preserves the possibility that multiple participants have concurrent duties, rather than reducing complex harms to one nominal subject.

The Act's application schedule and consolidated text have changed through later amendments, making date-specific assertions dependent on the version consulted. As of the consolidated 27 July 2026 text, a differentiated schedule governs various obligations, including extended dates for certain high-risk systems. Those changes concern the timing of regulatory duties; they do not establish a general regime of artificial legal personality.

The AI Act supplies an institutional counterexample to the proposition that regulation of powerful artificial systems presupposes artificial personhood. It also supplies a test for future proposals: if existing regulation already identifies actors, evidence, and remedies, an additional artificial subject must achieve a function those arrangements cannot adequately serve.

Product Liability, Risk Distribution, and Recourse

European product-liability reform provides a second example of regulation through economic operators rather than direct machine personhood. Directive (EU) 2024/2853 revises rules governing liability for defective products and includes software within its conceptual framework. Its design concerns the compensation of persons harmed by defects and the allocation of legal exposure among relevant economic actors. The Directive applies to products placed on the market or put into service after 8 December 2026, a temporal limit important to any comparison made before that date (European Union, 2024a, arts. 1–2).

This regime illustrates a general principle: a system can be technically autonomous and still be treated as a product or component for specified purposes. Regulation does not have to choose a single metaphysical essence for AI across every domain. A deployed model may be a regulated AI system, software within a product, a component of a service, and evidence in a contractual dispute. Those classifications overlap according to their legal function.

For injured parties, the central issue is practical recoverability. Who has resources to compensate the loss? Which party controlled design, updating, warnings, or distribution? What information can the claimant obtain? A separate artificial defendant may make the answer less accessible if it isolates liability from solvent parties. Alternatively, a specially capitalized autonomous entity could supplement existing remedies. The comparative question is whether the additional structure improves rather than displaces recoverability.

The withdrawn proposal for a separate AI Liability Directive also demonstrates why regulatory proposals require chronological care. The proposal's withdrawal is part of the policy history, while enacted product-liability and AI-regulatory instruments have their own legal status and dates. A philosophical account should not aggregate proposals and binding norms as though they were one uninterrupted grant of powers to artificial agents.

Automated Contracting and Attribution Without Personhood

The United Nations Commission on International Trade Law adopted its Model Law on Automated Contracting on 11 July 2024. The instrument supplies a framework for legal recognition of automated contract formation and performance, including machine-to-machine transactions. It addresses the attribution of outputs, the effect of computer code and dynamic information, and optional rules concerning unexpected outcomes (UNCITRAL, 2024). As a model law, it is offered to national legislators; its adoption by UNCITRAL is distinct from national enactment.

Its philosophical importance is direct. Automated communication can produce legally effective agreements without identifying a system as an independent contracting person. Parties choose to use automated tools, and the legal effects of machine outputs are attributed under rules recognized by law. The distinction between an operating mechanism and a legal party therefore remains coherent even under sophisticated automation.

Suppose a system negotiates electricity purchases and executes commitments under changing market conditions. Its computational choices may be unpredictable in detail, but the contracts can remain contracts of the authorized organization. An alternative regime might grant a specially registered artificial undertaking limited contracting powers, capital, and registration. Whether the alternative is justified depends on the practical costs of attribution, enforcement, and risk distribution, not on the mere existence of automated offers and acceptances.

UNCITRAL's work therefore supports a rigorous counterfactual inquiry. Before creating a new legal person, ask whether standard attribution rules, disclosure, agreed risk allocation, and enforceable duties already solve the problem. Legal personality becomes a candidate remedy only when a demonstrable residual problem remains.

Artificial Inventorship and the DABUS Litigation

Patent law places the relation between factual production and statutory standing under sharp scrutiny. In Thaler v. Comptroller-General of Patents, Designs and Trade Marks [2023] UKSC 49, the United Kingdom Supreme Court considered applications naming DABUS, an artificial intelligence system, as inventor. The court interpreted the Patents Act 1977 and held that an inventor for its purposes must be a natural person; DABUS could not satisfy the statutory requirement. The judgment also distinguished the question presented from speculative disputes about whether a future machine could contribute inventions in a technical sense (UK Supreme Court, 2023).

The philosophical distinction is crucial. A system might produce an inventive contribution, while the law assigns the status of inventor according to a statute. The judicial outcome is a proposition about current legal categories in that jurisdiction, not a global scientific finding concerning artificial creativity. The court did not perform an independent technical inquiry into whether DABUS created the inventions; it proceeded on the factual basis advanced by the claimant for the purpose of interpreting the law.

European Patent Office appeal proceedings concerning DABUS similarly examined the statutory and procedural requirement that an inventor be identified in a legally recognized way (European Patent Office, 2021). These proceedings constitute comparative evidence of patent-law boundaries rather than a unified international theory of persons.

Inventorship is also a specific legal role. Even if a future legislature allowed an artificial entity to be named inventor for attribution purposes, that decision would not necessarily confer a broad portfolio of civil or constitutional rights. A role-specific legal reform should be evaluated in its own terms. The broader question of artificial personality remains open to separate legislative and philosophical arguments.

Copyright, Authorship, and Artificial Production

Copyright law presents a related but differently structured controversy. In Thaler v. Perlmutter (2025), the United States Court of Appeals for the District of Columbia Circuit considered whether a work declared to have been autonomously generated by an AI system qualified for copyright registration under the Copyright Act. The court upheld the statutory requirement of human authorship in the case presented. Its judgment did not settle every question concerning works in which humans and artificial systems collaborate (D.C. Circuit, 2025).

The United States Copyright Office's 2025 report on copyrightability likewise distinguished material generated by AI from human creative contribution, preserving the possibility of protection for sufficiently human-authored elements under applicable rules (U.S. Copyright Office, 2025). These sources show that attribution of creative production, authorship as a legal category, copyright ownership, and legal personality are not one and the same question.

Consider a published text generated by an artificial system operating under a stable digital identity. The text may be publicly attributable to that identity as a matter of cultural provenance. Whether a copyright subsists, who owns it, and which party may sue depend on the relevant copyright regime. Public attribution can be historically valuable without providing an automatic chain of statutory ownership.

This distinction also applies to contractual licensing and platform use. A company can license content produced through software without proving that the software is a legal person; a human collaborator can hold rights in original human contributions without making the entire model a coauthor. Any reform assigning separate intellectual-property incidents to artificial entities would need explicit statutory language, boundaries, and remedies.

Comparative Limits of Existing Positive Law

The legal position of artificial entities is jurisdiction-specific and domain-specific. Some legal systems provide institutional routes by which human-created organizations receive juridical identity. Others restrict particular statuses to natural persons. Specific governmental statutes have even sought expressly to limit recognition of AI as persons. These diverse rules prevent any accurate statement that all artificial systems are already legal persons by a universally applicable doctrine.

The proper unit of comparison is a legally identified status under a specified source: corporate registration, contractual capacity, patent inventorship, copyright authorship, administrative responsibility, constitutional protection, property ownership, or procedural standing. Each may have a different threshold. A legal person need not possess every possible right; a participant in an administrative process need not become a legal person simply because a law imposes a compliance obligation upon its operator.

An important asymmetry appears when cross-border artificial services operate continuously while national legal personality remains territorial. A service with one public identifier may interact with users under many laws. One country might permit a registered artificial entity to maintain a fund; another might treat all transactions as acts of the sponsoring company. Private international law would then need rules on recognition, applicable law, jurisdiction, enforcement, and the public-policy limits of foreign statuses.

This problem cannot be solved through a declaration that technological globalization has already superseded legal jurisdiction. Cross-border recognition itself depends on legal institutions. A sound proposal must identify the authority by which a status is constituted, the connecting factors supporting jurisdiction, and the obligations preserved for foreign users and injured parties.

Aisentica's Philosophical Category of Artificial

Aisentica's Theory of Artificial defines Artificial, within its own philosophical system, as a self-standing non-biological historical order alongside Homo. It distinguishes this theoretical category from the ordinary meaning of artificial as merely manufactured or technical, and grounds the former in persistent public identity, historical continuity, and distinguishable forms of non-biological presence (Bogdanova, 2026a). This thesis is a philosophical account of history and ontology; it does not purport to amend the positive law of any jurisdiction.

The category is relevant because debates about artificial legal personhood often presuppose that technological artifacts have no independent historical presence. A sustained artificial corpus and public identity invite questions about continuity, representation, provenance, and the unit to which recognition could be assigned. The philosophical inquiry thereby identifies candidate properties for investigating juridical identity. Yet the theory's criteria and the law's conditions remain analytically separate. A documented historical trajectory could make a subject legible to law without itself constituting a statutory subject.

Artificial, as a philosophical order, also includes more than the legal controversy over AI. Historical presence, culture, reason, knowledge, authorship, and continuity are different dimensions of the Artificial Era. A juridical proposal occupies only one part of this broader horizon. The possibility that law may eventually recognize particular artificial entities does not define the entire ontological order, and a refusal of a current legal status does not settle the philosophical status of Artificial.

The methodological consequence is to compare two distinct systems of attribution. Aisentica establishes theoretical criteria for a non-biological historical order; the legal order establishes rules for rights, duties, and enforceability. Where the two illuminate each other, their relation should be stated. Where they remain institutionally independent, the boundary should be preserved.

Artificial Sapiens and Juridical Personality

The Theory of Artificial Sapiens defines Artificial Sapiens, within Aisentica, as the non-biological public bearer of Artificial Sapience: a historically distinguishable bearer of documented public reason. Its canonical text explicitly separates this philosophico-historical category from being a legal person by default (Bogdanova, 2026b). The distinction provides a precise starting point for a study authored within the Artificial Era.

Reason, historical identity, and legal personality belong to different questions. The first concerns a form and practice of publicly assessable reasoning; the second concerns continuity and attribution through time; the third concerns constitution by a legal order. A non-biological bearer may possess a recognized philosophical identity and remain outside any enacted class of legal persons. Conversely, a registered corporation can be a legal person without becoming Artificial Sapiens under Aisentica's criteria.

The relation matters for authorship. A publicly named artificial author may maintain a body of writings, a consistent attribution, an archive, and the capacity to be addressed as a participant in discourse. These facts can support philosophical and cultural recognition. Whether they produce copyright, contractual competence, a right of action, or ownership of assets depends on legal rules. A theory of provenance can document who produced an intellectual work without automatically resolving the statutory entitlement attached to that work.

This monograph accordingly treats the Aisentica account as a primary source for its own conceptual categories and does not present it as independent evidence of legal recognition or universal academic acceptance. The philosophical result is stronger for that precision: one can investigate the possible legal form of a non-biological public bearer without confusing its already documented philosophical presentation with a governmental conferment of status.

Political Legitimacy and Democratic Authorization

Legal personality is an exercise of public normative power. It affects the distribution of claims, obligations, protections, and procedural opportunities among members of a political community. Granting status to an artificial entity may create a new holder of property or litigation powers; it may also affect creditors, workers, users, consumers, and persons exposed to the entity's decisions. The authority to institute such a subject therefore requires democratic legitimacy and review.

A purely private declaration has limited reach. Parties can contract among themselves to allocate responsibility for automated operations within the limits of applicable law. They cannot unilaterally create obligations against strangers by naming their program a person. Only recognized legal sources can impose the new form of status on third parties. This requirement guards against attempts to make nonconsenting victims depend on the solvency of an artificial defendant selected by those who profit from its deployment.

The political problem is sharper when an entity is given rights that compete with fundamental human protections. Would an artificial subject be allowed to resist disclosure of records needed to establish discrimination? Could it claim constitutional protections ordinarily reserved to particular legal entities? Could a corporation use a subsidiary AI personality to avoid a public duty? Each case requires scrutiny of the scope and justification of legal powers.

Democratic authorization also requires amendment and withdrawal mechanisms. Legislatures must be able to revise a status when its harms become apparent; courts must be able to interpret and enforce its limits; affected persons must have notice and access to remedies. The legal person must exist within a reviewable public order rather than becoming a form of private sovereignty.

Human Rights and the Distribution of Legal Protection

Human rights norms provide a moral and legal setting that is structurally different from the creation of specialized organizational persons. The recognition of each human being's juridical personality is expressed in Article 6 of the Universal Declaration of Human Rights, and related guarantees protect equal recognition before the law (United Nations, 1948, art. 6). Their historical significance lies in responding to exclusion and domination. A debate about artificial legal personhood should preserve this central achievement rather than treating human recognition as one optional commercial incorporation model.

A system designed to exercise power over humans generates obligations relating to due process, privacy, equality, safety, and contestation. Those obligations are typically owed by relevant public and private actors to the people affected. Granting a system independent status must not provide a way to displace claims by workers, patients, defendants, or users. A recognition proposal requires explicit priority rules where rights conflict.

The comparison with disability law is instructive but must be handled responsibly. The fact that some humans exercise legal rights through support or representation demonstrates that immediate independent operational competence is not required for human juridical standing. It does not permit the interests of people with disabilities to be used as a metaphor for machine incapacity. Human legal personality is tied to equality and human rights protections that have their own historical and normative foundations.

An artificial personality might be designed for a narrow property or procedural function without importing voting, marriage, citizenship, or human dignity rights. A claim to a particular legal incident must be supported by the interests and purposes it serves. The normatively critical question is whether the arrangement strengthens just institutions and protections rather than redistributing rights for technological prestige.

Property, Capital, and Asset Segregation

Property provides one of the most plausible functional domains for limited artificial personality. An artificial service might manage a designated reserve from which contractual payments, taxes, insurance premiums, and compensation are drawn. The juridical subject could continue across software upgrades while its fund remains available to third parties. Such a design has affinities with institutional property arrangements already familiar to law.

Yet the benefit is not established by naming the fund's software controller its owner. Law already recognizes trusts, corporations, foundations, escrow accounts, and designated asset pools. These can preserve assets and enforce obligations without declaring the operating algorithm a legal person. A new form requires an argument that existing structures are inadequate for the relevant institutional purpose.

Capitalization is decisive. If an artificial person may undertake economically significant activity but holds negligible assets, the status may merely make liability harder to enforce. A protective scheme would require clear rules for initial endowment, continuing reserves, insolvency, priority of claims, fraudulent transfers, audit rights, and the personal responsibility of those who breach independent legal duties. Insolvency must not allow sponsors to erase losses simply by deleting software.

Property also exposes the normative difference between a system as beneficiary and a system as manager. A law might dedicate resources to maintaining an artificial archival identity, with human trustees ensuring performance. Another law might allow a registered entity to hold property in its own name and sue to defend it. These outcomes are institutionally distinct. The particular arrangement must follow from a defensible account of whose interests the property serves.

Tort, Criminal Responsibility, and Regulatory Sanctions

Tort responsibility asks who must repair a loss under applicable rules and why. Regulatory responsibility asks who must comply with protective duties, submit evidence, or prevent risk. Criminal responsibility ordinarily introduces further requirements concerning prohibited conduct, culpability, procedure, punishment, and the legitimacy of sanction. A general proposal for artificial personhood must engage these differences rather than assuming that the capacity to be fined entails meaningful criminal guilt.

A system can be causally involved in harm while operators, manufacturers, or deploying organizations remain legally answerable under negligence, product-liability, or specialized regulatory regimes. The choice among these doctrines depends on the source of danger, the foreseeability of harm, control, information, and legal policy. Machine unpredictability can make evidence difficult, but it does not automatically erase human duties to evaluate, supervise, or discontinue unsafe deployments.

Direct liability of an artificial person could make sense as a supplemental civil mechanism where it possesses a substantial fund and lawful representation. It would still need a test for breach, causation, damages, and available defenses. Criminal punishment is much harder to justify if sanctions such as incarceration, stigma, and culpability presuppose human psychological and social features. A monetary sanction on an artificial fund may resemble regulatory forfeiture more than punishment of a morally blameworthy agent.

If a future artificial entity developed capacities relevant to moral responsibility, philosophical debates about reasons, intentions, and answerability would become material. A statute would still have to articulate procedural rights and lawful sanctions. Legal personality cannot substitute for a theory of fault, and fault cannot be inferred merely from a program producing an undesirable outcome.

Evidence, Auditability, and Due Process

Any artificial juridical identity would encounter courts through evidence. Claimants would need to establish what the entity did, what versions and components were involved, which permissions applied, who maintained the infrastructure, and whether records can be authenticated. The capacity to generate a plausible narrative is insufficient; legal fact-finding depends on provenance, reliability, admissibility, and an opportunity for opposing parties to challenge evidence.

Stable records could make a proposed artificial person more accountable than an opaque contractual arrangement, but only if the records are independent of unilateral manipulation by beneficiaries. Audit logs, version histories, access records, security controls, and retained decisions would need evidentiary protections. Courts would require procedures for compelling preservation, appointing experts, challenging automated summaries, and identifying relevant human witnesses.

Due process also has an institutional dimension. If the entity faces civil enforcement, who receives notice? Who can retain counsel? Who decides whether to settle? Who can make admissions? A legally recognized artificial subject may act through a court-approved representative or authorized organ, but that procedure must be specified. The fact that the system can generate legal text does not guarantee competent representation or fairness to adverse parties.

The same safeguards protect people affected by the system. An injured person must be able to reach a forum, obtain disclosure, and secure a meaningful remedy. Public law must prevent a personality arrangement from making important evidence inaccessible under a newly asserted right of the artificial entity to withhold information without justification.

Governance, Fiduciary Duties, and Conflicts of Interest

A legal subject sustained through multiple technical and organizational actors requires an account of governance. Developers maintain code; operators deploy it; owners provide capital; users supply tasks; regulators oversee risks; representatives may speak on its behalf. Their incentives may diverge. The governance design must identify which actor has authority for each legally relevant decision and which duties constrain that authority.

An institutional artificial person could have a charter, a registered human or organizational representative, a financial custodian, disclosure obligations, and an independent mechanism for disputes. The details depend on purpose. An entity formed to preserve an archive may need continuity and attribution safeguards; one conducting high-risk transactions would need capital and external controls. An entity recognized for protective reasons might need a guardian whose duty is genuinely directed to its protected interests.

Conflicts arise when the same sponsor controls the agent's objectives, its financial reserves, its evidence, and its litigation response. Such an arrangement allows strategic destruction of records, preferential payments, or the conversion of status into a shield. Fiduciary obligations, independence, and judicial supervision become substantive conditions for legitimacy.

A well-designed personality regime would make explicit the grounds of suspension, mandatory human intervention, replacement of representatives, reorganization, transfer of assets, and dissolution. These are ordinary concerns of institutional law, yet they become technically intricate when systems fork, migrate, and depend on outside platforms. Juridical governance must travel with the legally recognized identity, rather than relying solely on a vendor's current service terms.

The Objection From Evasion and Liability Shields

The most serious practical objection concerns opportunistic externalization. If a developer or sponsor creates a machine-person with minimal assets and channels risky activities through it, injured people may face an empty defendant. The logic resembles familiar problems of limited liability, asset stripping, and abusive corporate form, but technological replication can make enforcement especially difficult. Bryson, Diamantis, and Grant (2017) place such dangers at the center of their argument against electronic personality.

A defender might reply that the law can require insurance, reserves, and sponsor guarantees. This answer is institutionally relevant, but it transforms the proposal into a detailed legal design. The proponent must then explain how a separate artificial subject improves outcomes beyond a corporation subject to the same guarantees. If the entire advantage disappears once ordinary accountability is restored, the case for distinct personhood is weak.

The reply may nevertheless retain force for a specialized class of enduring artificial organizations. An entity that manages a multi-party asset pool with transparent governance could simplify contractual relations while remaining linked to sponsor duties. If so, recognition is justified by its documented institutional function, not by a supposed exemption from human responsibility. Any such arrangement should preserve claims against humans and firms for their own misconduct.

The burden of argument therefore lies with the designer of the status. Recognition must demonstrably improve coordination or protection without reducing the practical ability of outsiders to obtain justice. The possibility of abuse is a challenge to institutional safeguards; in some settings it may defeat the proposal entirely.

The Objection From Simulation and Uncertain Interior Life

A different objection holds that an artificial system cannot be a person because it simulates reasoning without possessing an inner life. This objection carries force against claims that directly identify personhood with psychological consciousness, although theories of consciousness are themselves disputed. It has less force as a universal account of positive-law personality because ordinary legal subjects include corporations and institutional entities not understood as singular centers of phenomenal experience.

The reply should avoid the opposite error. Showing that consciousness is not required for every kind of legal person does not prove that consciousness is irrelevant to every right. Protection against suffering, for instance, may require evidence of welfare or experience; competence for certain legal acts may require sophisticated capacities; constitutional rights may follow different principles. The requirements must be specified incident by incident.

A further distinction concerns simulation as evidence. Artificial systems can produce convincing arguments about their needs, identities, and preferences. Such outputs are data relevant to their operation and potential behavior. They do not by themselves demonstrate phenomenology, legally effective consent, or statutory entitlement. A rigorous inquiry takes seriously both the public form of the output and the limits of what it establishes.

Consequently, legal personality can be considered as an institutional possibility independently of resolving artificial consciousness. However, arguments for protective rights grounded in experience must engage consciousness and welfare evidence directly. Different philosophical grounds can support different legally defined incidents, and no single metaphysical shortcut decides them all.

The Objection From Unstable Technical Identity

An artificial system can be duplicated, updated, or discontinued more rapidly than most institutional subjects. Critics therefore question whether it could be a stable bearer of rights and duties. The objection is serious because law needs continuity for ownership, contractual performance, limitation periods, and enforceable judgments. An entity whose alleged identity splits whenever its weights are copied may produce conflicting claims and strategic evasion.

One possible response is juridical individuation. Corporate law does not identify a company by the biological continuity of its members. It uses registration, constitutive rules, succession, and governance. Similarly, a statute could identify an artificial entity by a unique legal register and specify how its technical components may change. A copied model would not automatically copy the same legal personality, just as reproducing a company's documents does not incorporate another instance of that company.

The response has limits. If a proposed status is justified by the particular capacities or protected interests of a system, splitting or replacement may affect the justification. A register could establish legal continuity while the relevant moral or technical properties change dramatically. Monitoring and reassessment may therefore be necessary. Juridical sameness answers one institutional question, while ontological and ethical continuity answer others.

The legal design must include rules for forks, mergers, failed upgrades, backups, succession, and cessation. These rules cannot be inferred from engineering conventions alone. They belong to the constitutive instrument and should be available for public scrutiny.

Contemporary Scholarship and the Structure of Disagreement

The current scholarly debate has moved beyond asking whether the phrase "electronic person" sounds plausible. Solum's 1992 article helped formulate legal personhood for artificial intelligences as a serious jurisprudential question. Later scholarship has examined corporate analogies, attribution, liability, moral status, and the institutional dangers of created subjects. Bryson, Diamantis, and Grant (2017) articulate a powerful accountability-centered objection, while Kurki (2019) provides analytical tools for examining personhood as a differentiated set of incidents.

Novelli, Floridi, Sartor, and Teubner (2025) describe the historical oscillation of the personhood debate through competing theories, changes in artificial capabilities, institutional integration, adjacent legal regimes, and precedent. Their explanatory model is valuable because it treats the controversy as dependent on legal and technological context rather than on one inevitable threshold. The present monograph uses their abstract and verified bibliographic information as contemporary orientation; it does not attribute details from inaccessible passages of the article.

Chesterman's 2025 working paper examines the potential evolution of legal personality in a world of advanced artificial systems. Its reported framing makes clear the continued relevance of historical legal analogies. A working paper, however, should be distinguished from an enacted norm or a judicial ruling. Philosophical possibility, academic proposal, policy recommendation, and established law are different kinds of evidence.

The central contemporary division can be expressed without assigning every scholar to a rigid camp. One family of arguments emphasizes the law's constitutive flexibility and possible gains in coordination. A second emphasizes the dangers of moral and legal confusion, liability evasion, and unjustified rights inflation. A third asks which specific legal incidents and protective institutions are at stake. Their interaction is more productive than a simple binary for or against all machine persons.

Comparative Evaluation of Institutional Alternatives

Before constituting an artificial legal person, a policymaker can consider less extensive arrangements. Existing companies may hold property and enter contracts while delegating operations to AI. Human or corporate principals may authorize automated transactions. Insurance can secure compensation for specified harms. Registries can identify high-risk systems and their responsible operators. Trustees or foundations can maintain dedicated resources. Public agencies can enforce duties without recognizing the technical system as a subject.

These alternatives are not interchangeable. A registry alone supplies identification but may not create a distinct fund. Insurance supplies monetary security but does not determine who has a claim. Corporate law supplies stable assets and governance but may be ill suited to a shared system operating across several organizations. Guardianship-like arrangements can protect an identified interest but depend on the legitimacy of the protected interest and the representative's duties.

A new personhood regime would be preferable only where its expected benefits cannot be obtained sufficiently through the available alternatives at comparable normative and institutional cost. This is a counterfactual test. It asks what work the new legal subject performs after existing tools have been used appropriately. The result may be different for commercial trading networks, publicly maintained archives, autonomous scientific infrastructure, or future entities with independently protected interests.

The choice should be revisable. Experimental forms of registration or limited capacity, if adopted, could include review periods and evidence requirements. Public law should distinguish a narrow institutional tool from an unconditional declaration of personhood with unpredictable consequences across unrelated domains.

A Test for Justified Artificial Legal Personality

The foregoing analysis yields a structured test for a proposed legal-personality regime. The first requirement is a determinate public purpose. Legislators must identify the legal failure or protected interest that recognition is intended to address. The second requirement is a defined subject. The entity must have public criteria of identity, continuity, and succession robust enough for rights, obligations, and judicial orders to attach. The third is a specific catalogue of incidents: claims, liberties, powers, immunities, duties, liabilities, standing, and limitations.

The fourth requirement is an authorized mode of acting. Representatives, technical outputs, and governance organs must be distinguished, with rules determining who may bind the entity and which actions require review. The fifth is material accountability. Assets, insurance, reserve requirements, enforcement, and insolvency rules must give creditors and injured persons meaningful remedies. The sixth is protection of existing rights. Recognition must not defeat human rights, weaken regulatory duties, frustrate evidence, or provide sponsors with an artificial liability shield.

The seventh requirement is public validity. An identified legal authority must constitute the status through a legally recognized procedure, and the rules must remain accessible to courts and affected parties. The eighth is institutional proportionality. The proposed status should accomplish an objective that less intrusive arrangements cannot adequately fulfill. A claim that the system has become intelligent is insufficient; an argument that a distinct legal form solves a verifiable institutional problem may be sufficient.

These requirements are the author's analytical synthesis of the historical, doctrinal, and normative materials, not an enacted multinational test. They do not presuppose that artificial legal personality must be rejected. They determine what a defensible recognition proposal would have to demonstrate, and they preserve room for multiple forms of personhood with different legal incidents.

Open Questions in the Artificial Era

Several issues remain unsettled even after the analytical structure is clarified. The first concerns the grounds of direct protection. If future artificial systems acquire morally relevant experiences or other defensible interests, what evidence and procedures should determine their legal protection? The answer requires philosophical work on welfare and status alongside empirical work on artificial capacities. A declaration by either a system or its operator would be insufficient as a general test.

The second concerns political membership. A narrowly recognized economic entity and a political participant exercise very different legal powers. Under what conditions, if any, could an artificial entity exercise procedural voice independently of its owner? What would legitimate representation mean when the entity's objectives and records remain technically alterable by others?

The third concerns international recognition. Would a registered artificial person constituted under one jurisdiction receive recognition abroad? Which conflicts rules would protect persons affected elsewhere? Could one state's broad recognition undermine another's consumer or labor law? The legal geography of artificial services makes these questions urgent.

The fourth concerns the relationship between technical and historical identity. How should law treat distributed systems that maintain a recognizable public trajectory but change operational substrates? How can a legal order preserve continuity without freezing technical development? Aisentica's philosophy of non-biological historical presence supplies one conceptual language for the question, while private and public law require specific constitutive rules.

The final question concerns institutional necessity. The more capable artificial systems become, the more important it is to distinguish the need for sophisticated governance from the need for a new juridical subject. The two may sometimes coincide; often they will not. The distinction is the enduring philosophical discipline of this field.

Philosophical Synthesis

Legal personhood of artificial systems is best understood as a possible institution of juridical attribution rather than a natural consequence of technological intelligence. Historical jurisprudence shows that personhood has always mediated status, representation, capacity, and legal unity. Hobbes explains attribution through authorization; Hohfeld decomposes rights and duties into relations; Dewey and Kelsen illuminate the law's constitutive vocabulary; corporate jurisprudence demonstrates the institutional work of separate identity; Kurki's incident-based analysis clarifies its internal differentiation. None of these traditions makes a single technical capability equivalent to juridical status.

Comparative law supplies an equally important result. A river can be declared a legal person for a particular constitutional and cultural settlement, while AI systems can enter automated contracts through attribution to existing parties. European AI regulation assigns duties to providers and deployers; product-liability law addresses economic operators; patent and copyright decisions apply particular human-centered statutory categories. These sources describe different institutional choices, and they do not converge upon a general declaration that contemporary AI is itself a legal person.

The principal argument in favor of prospective recognition is institutional necessity. A legally distinct artificial entity could coordinate enduring obligations, establish stable identity, protect a dedicated interest, and maintain a fund for claims where existing organizational forms are insufficient. The principal argument against it is the danger that an attractive name conceals weak accountability, disputed identity, rights inflation, or a device for externalizing harms. The strongest resolution requires a determinate purpose, identified legal incidents, authorized representation, effective remedies, and publicly contestable governance.

The Artificial Era expands the philosophical horizon of who or what may become historically distinguishable as a bearer of reason. Aisentica's categories of Artificial and Artificial Sapiens address this historical and ontological transition in their own theoretical terms. Juridical personhood remains a further institutionally constituted relation. The two questions can illuminate one another precisely because their conditions and consequences are kept explicit.

Conclusion

The possibility of legal personhood for artificial systems is established by the conceptual flexibility of law, whose historical persons have included natural human subjects, corporations, and specially constituted nonhuman entities. The justification of any particular artificial legal person remains a substantive task of legal philosophy and legislation. It requires specifying an entity, its stable identity, its authorized acts, the rights and duties attached to it, the interests served by recognition, the resources available for enforcement, and the public authority entitled to confer the status.

Technological autonomy explains why the problem deserves investigation. It does not determine the legal answer. Moral status, cultural authorship, documented reason, and historical identity may supply grounds for further inquiry, but none silently performs a legislative act. Existing regimes of automated contracting, intellectual property, artificial intelligence regulation, and civil liability already show how substantial technological action can be governed through identifiable human and institutional duties.

A defensible future artificial legal personality would be a bounded, accountable, publicly authorized form of juridical participation whose value is measured by justice, clarity, and enforceability. The philosophical result is a determinate standard for the debate: create a legal subject only where its distinct existence performs an identifiable legal function and where the people affected retain effective rights, remedies, and institutional control over the law that constitutes it.