Skip to main content

The Implications of Modern Business–Entity Law for the Regulation of Autonomous Systems

  • Shawn Bayern

Nonhuman autonomous systems are not legal persons under current law. The history of organizational law, however, demonstrates that agreements can, with increasing degrees of autonomy, direct the actions of legal persons. Agreements are isomorphic with algorithms; that is, a legally enforceable agreement can give legal effect to the arbitrary discernible states of an algorithm or other process. As a result, autonomous systems may end up being able, at least, to emulate many of the private–law rights of legal persons. This essay demonstrates a technique by which this is possible by means of limited liability companies (LLCs), a very flexible modern type of business organization. The techniques that this essay describes are not just futuristic possibilities; as this essay argues, they are already possible under current law.

Hide All

1 A somewhat more formal definition that conveys a similar message is that a legal person, in the sense I mean it in this paper, is anything to which the law can ascribe any Hohfeldian jural relation, such as a right, duty, or power. See Hohfeld, Wesley N., “Fundamental Legal Conceptions as Applied in Judicial Reasoning”, 26 Yale L.J. (1917) pp. 710 et sqq. (defining and classifying “jural relations”).

2 See Pollman, Elizabeth, “Reconceiving Corporate Personhood”, 2011 Utah L. Rev. pp. 1629 et sqq., at pp. 1629–32 (discussing a similar distinction in the meaning of corporate personhood and applying it to modern debates about constitutional rights). For a recent discussion of different functions of corporate personhood generally, see Blair, Margaret M., “Corporate Personhood and the Corporate Persona”, 2013 U. Ill. L. Rev. pp. 785 et sqq.; see also Orts, Eric W., Business Persons: A Legal Theory of the Firm (Oxford: Oxford University Press, 2013).

3 Of course, nobody thinks two corporations can get married, even though they are undisputedly legal persons in the realm of private law and organizational law. (Conversely, nobody thinks two natural persons can merge.)

4 See Restatement (Third) of Agency Law § 1.04 cmt. e (2006)(” [A] computer program is not capable of acting as a principal or an agent as defined by the common law. At present, computer programs are instrumentalities of the persons who use them. If a program malfunctions even in ways unanticipated by its designer or user, the legal consequences for the person who uses it are no different than the consequences stemming from the malfunction of any other type of instrumentality.”).

5 Bayern, Shawn J., Closely Held Organizations (Durham, North Carolina: Carolina Academic Press, 2014) at pp.9394 (describing ways in which organizational law serves as an interface to other areas of private law).

6 See Del. Ltd. Liab. Co. Act. § 18–215.

7 For more discussion of this type of autonomous entity, see Bayern, Shawn, “Of Bitcoins, Independently Wealthy Software, and the Zero–Member LLC”, 108 Nw. U.L. Rev. (2014) pp. 1485 et sqq., at p. 1486 (“Bitcoin allows autonomously operating software—such as a computer virus or the software that manages a network of vending machines—to exercise control over significant wealth, not as an intermediary for individuals or companies, but rather, in a functionally meaningful sense, in its own right.”).

8 See Turing, A.M., “Computing Machinery and Intelligence”, 59 Mind (1950) pp. 433 et sqq. (proposing that a functional answer to the question “Can machines think?” be determined through an “imitation game” in which a computer seeks to impersonate a human in written correspondence).

9 The prominence of the bridge–building example in American commentary may date to the case Charles River Bridge v. Warren Bridge, 36 U.S. 420, at p. 420 (1837) (involving “a corporation created by an act of the legislature of the state of Massachusetts, passed on the 9th of March, 1785, entitled ‘An act for incorporating certain persons for the purpose of building a bridge over Charles river, between Boston and Charlestown, and supporting the same during forty years.’”).

10 See Bayern, Closely Held Organizations, supra, note 5, at pp. 209–10. 11 See Melvin Aron Eisenberg, “Legal Models of Management Structure in the Modern Corporation: Officers, Directors, and Accountants”, 63 Calif. L. Rev. (1975) pp. 375 et sqq., at p. 376 (“Instead [of a managerial board], in small, closely–held corporations the business is typically managed directly by owner–managers, while in large, publicly–held corporations … the business is typically managed by the top executives.”).

12 Del. Gen. Corp. L. § 141(b).

13 E.g., Model Bus. Corp. Act. § 8.03(a) (“A board of directors must consist of one or more individuals … .”); ibid. § 1.40(13) (“‘Individual’ means a natural person.”).

14 See ibid. § 7.32 cmt.

15 Ibid.

16 See ibid.

17 Ibid. § 7.32(a).

18 See supra, note 8.

19 The term “shareholder agreement” as used in the MBCA appears to cover single-party operating agreements; at least, nothing in the Act equates “agreement” with “contract” or requires the assent of two or more parties. Even if such a requirement existed, however, it would not change much of the discussion in the text.

20 Though the Model Business Corporation Act is not fully clear on this point, if E is the sole founding shareholder and remains the sole shareholder, E can likely revoke the operating agreement. See MBCA § 7.31(b) (“An agreement authorized by this section shall be … subject to amendment only by all persons who are shareholders at the time of the amendment, unless the agreement provides otherwise … .”).

21 Ibid. § 6.01(b) (“The articles of incorporation must authorize … one or more classes or series of shares that together have unlimited voting rights … .”). Note that § 7.32(a) of the Act does not include eliminating the notion of shareholders from its list of the capabilities of an enforceable shareholder agreement.

22 E.g., ibid. § 1.40(21) (“‘Shareholder’ means the person in whose name shares are registered … .” (emphasis added)).

23 Melvin Aron Eisenberg & James D. Cox, Business Organizations: Cases and Materials, 11th ed. (St. Paul, MN: Foundation Press, 2014), at p. 132. Cf. supra, note 1 regarding Hohfeldian jural relations.

24 Unif. P’Ship Act. § 25(2)(a) (1914) (creating and defining “tenancy in partnership”).

25 RUPA § 202(a) (“[T]he association of two or more persons to carry on as co–owners a business for profit forms a partnership”).

26 RUPA § 103(a) (providing that “relations among the partners and between the partners and the partnership are governed by the partnership agreement” except for a statutorily enumerated list of specific prohibitions).

27 See RUPA § 103(b)(8).

28 See Robert W. Hillman & Donald J. Weidner, “Partners Without Partners: The Legal Status of Single Person Partnerships”, 17 Fordham J. Corp. & Fin. L. (2012), pp. 449 et sqq. (presenting both sides of the debate). 29 LLCs provide other advantages over general partnerships, such as limited liability, and if nothing else they require only one (rather than two) legal persons to establish them. The same analysis as in the text applies to limited liability partnerships (LLPs) governed by RUPA. Broadly similar analysis applies to limited partnerships governed by various versions of the Uniform Limited Partnership Act, but for ease of exposition I leave limited partnerships outside the scope of this article.

30 LLCs have two common organizational paths, “member–managed” and “manager-managed.” See, e.g., Unif. Ltd. Liab. Co. Act § 407 (2006) [hereinafter “RULLCA”]; Bayern, Closely Held Organizations, supra, note 5, at pp. 243–45. The technique I describe in the text can achieve similar results with both types of LLCs; I focus on member-managed LLCs for simplicity.

31 RULLCA § 407(b) (“In a member-managed limited liability company, … [t]he management and conduct of the company are vested in the members.”).

32 Similarly, the individual may transfer any relevant intellectual property concerning the autonomous system to the LLC.

33 I have previously described an application of this technique in Bayern, “Of Bitcoins”, supra, note 7, at pp. 1495–98.

34 The parents may desire this structure (compared to one in which the children are members) as a convenient way to avoid having any legal obligations within the company to their children during their lifetime.

35 RULLCA § 701(a)(3) (“A limited liability company is dissolved, and its activities must be wound up, upon … the passage of 90 consecutive days during which the company has no members.”).

36 See supra text accompanying notes 17–18.

37 Bayern, “Of Bitcoins”, supra, note 7, at p. 1497.

38 RULLCA § 701 cmt.

39 Ibid.

40 RULLCA § 701. 41 RULLCA § 708(b). RULLCA uses similar language to apply to predissolution distributions, which an operating agreement can just as easily and uncontroversially override: “Any distributions made by a limited liability company before its dissolution and winding up must be in equal shares among members and dissociated members.” Ibid. § 404(a).

42 N.Y. LLC. LAW § 701(a)(1)(4) (emphasis added).

43 On a larger scale, the LLC may have employees who continue operating the LLC without any members. The employees’ powers would be determined ultimately by the operating agreement. It would not be unusual, for example, for a financially significant LLC in New York with a few employees to continue operating the entity normally while the operating agreement's process for determining new members proceeded separately, unaffected by (and not directly affecting) any operational concerns. If this situation arose in the context of a “family LLC” used by a wealthy family for the purposes of estate planning, I suspect the basic operation of the LLC, during its memberless period, would be uncontroversial except for substantive ambiguities in the operating agreement.

44 Moreover, as I have previously argued, “The permission of just a single state would be sufficient to enable autonomous businesses. An organizer of such a business merely would need to select the organizational law of a state that permits a perpetual autonomous LLC.” Bayern, “Of Bitcoins”, supra, note 7, at 1497. To act in other states, the business would ordinarily need to register as a foreign LLC, but (1) this step is often neglected with minimal results apart from limiting the ability of the entity to file a lawsuit in state court until the LLC corrects the technical registration defect, and (2) in any event, registration of a foreign LLC does not ordinarily alter its internal affairs, such as membership status or rights. This is a result of the general internal–affairs doctrine, a conflict–of–law rule that causes courts to defer on organizational matters to an entity's state of organizations. See the case Edgar v. Mite Corp., 457 U.S. 624, at p. 645 (1982) (“The internal affairs doctrine is a conflict of laws principle which recognizes that only one State should have the authority to regulate a corporation's internal affairs.”).A further side note: Even under an LLC act that does not permit memberless entities, the theoretical—and certainly the practical—possibility of entity cross-ownership enables very similar possibilities. The proposed technique is as follows: (1) Existing person P establishes member-managed LLCs A and B, with identical operating agreements both providing that the entity is controlled by an autonomous system that is not a preexisting legal person; (2) P causes A to be admitted as a member of B and B to be admitted as a member of A; (3) P withdraws from both entities. The result does not trigger the law's response to memberless entities, because what remains are simply two entities with one member each. Corporate statutes often have formal provisions that prevent this sort of cross–ownership from functioning successfully in corporations—at least as concerns the voting rights of shares—but there do not appear to be similar restrictions on LLCs, which of course in general provide for greater flexibility in arrangements of control and organization. Cf. MBCA § 7.21(b) (“Absent special circumstances, the shares of a corporation are not entitled to vote if they are owned, directly or indirectly, by a second corporation … and the first corporation owns, directly or indirectly, a majority of the shares entitled to vote for directors of the second corporation.”).

45 See, e.g., Brenner, Susan W., “Humans And Humans+: Technological Enhancement and Criminal Responsibility”, 19 B.U. J. Sci. & Tech. L. (2013) pp. 215 et sqq., at p. 285 (“I suspect [the law's concern with “how to enforce basic fairness and morality between a mix of human beings with varying abilities and also, perhaps, intelligent robots, cyborgs, chimeras, animals and alien beings”] will have to change even more if and when law decides to admit ‘objects’ (e.g., robots), animals (enhanced or not), semihumans (cyborgs and chimeras) or space aliens to the ‘legal person’ club currently monopolized by Standard human beings.”); Beard, Jack M., “Autonomous Weapons and Human Responsibilities”, 45 Geo. J. Int’l L. (2014) pp. 617 et sqq., at p. 663 (at least given current levels of technology, “[h]olding a robot accountable as a ‘legal’ person for war crimes as if it were a human appears to be impractical on many levels.”); Koops, Bert–Jaap, Hildebrandt, Mireille, and Jaquet–Chiffelle, David–Olivier , “Bridging the Accountability Gap: Rights for New Entities in the Information Society?”, 11 Minn. J.L. Sci. & Tech. (2010) pp. 497 et sqq., at p. 511 (“Depending on how novel legal persons are introduced, they could, in fact, destabilize familiar notions of responsibility that form the moral core of the law, reinforcing undesirable affordances of an increasingly independent technological infrastructure.”); Solum, Lawrence B., “Legal Personhood For Artificial Intelligences”, 70 N.C.L. Rev. (1992) pp. 1231 et sqq., at p. 1243 (“How then should the law answer the question whether an AI can become a legal person and serve as a trustee? The first inquiry, I should think, would be whether the AI is competent to administer the trust.”).

46 See RUPA § 202(a) (“[T]he association of two or more persons to carry on as co-owners a business for profit forms a partnership, whether or not the persons intend to form a partnership.”); see also Bayern, , “Three Problems (and Two Solutions) in the Law of Partnership Formation”, 49 U. Mich. J. L. Ref. (forthcoming 2016) (discussing the “accidental” formation of partnerships).

47 See Del. Ltd. Liab. Co. Act § 18-215(a) (“A limited liability company agreement may establish or provide for the establishment of 1 or more designated series of members, managers, limited liability company interests or assets. Any such series may have separate rights, powers or duties with respect to specified property or obligations of the limited liability company … and any such series may have a separate business purpose or investment objective.”). Indeed, the essence of series LLCs may well be the ability to create a multitude of legal entities without direct, ongoing interaction with the government. Under general partnership law, even though conceptually the same partners might create a multitude of accidental “series partnerships,” courts do not seem to have recognized that possibility. Nothing in RUPA prevents this possibility, but because general partnerships do not confer limited liability, there seems to be little reason for partners to argue that they have created a multitude of distinct, transaction–specific general partnerships.

48 See supra, text accompanying notes 2–5.

49 Trusts have long been permitted to aid impersonal beneficiaries, as in a “pet trust” that provides for the care of an animal. See Kim Bressant–Kibwe, “Pet Trust Primer”, November 2015, available on the internet at <> (last accessed on 3 December 2015). This type of trust has existed for hundreds of years. For example, Alexander Pope wrote of wealthy people who, finding little personal use for their money, “Die, and endow a College or a Cat.” Pope, Alexander, “Epistle III to Allen Lord Bathurst: Of the Use of Riches”, in John Butt (ed.) The Poems of Alexander Pope, pp. 570 et sqq., at p. 574 (New Haven: Yale University Press, 1963; original publication in 1733).

50 Similarly, for example, modern LLC law is probably flexible enough to create an entity with several members, none of which has any power—because the organizational document gives them none or requires complex checks and balances that are practically impossible to meet. Such an entity could functionally enable autonomous systems as well. It is important to recognize that nothing in the mandatory structure of LLCs requires that members be the economic beneficiaries of the entity, that members have any significant power, and so on; the structure of an LLC is flexible enough not to require any such traditional patterns.More generally, this line of thinking demonstrates that a robot can achieve significant interaction with the legal system merely with the consent of (or even merely with the absence of objections by) a passive, existing legal person.

51 Peter Steiner, [Cartoon], New Yorker, 5 July 1993, at p. 61.

52 George Edward Moore, Principia Ethica, rev. ed. (Cambridge: Cambridge University Press, 1993) at pp. 63–65 (discussing the “naturalistic fallacy”).

53 See supra, Part II.2.

54 See, e.g., Dan–Cohen, Meir, Rights, Persons, and Organizations (Berkeley: University of California Press, 1986) at p. 46 ; see also Wai, Katsuhito, “Persons, Things and Corporations: The Corporate Personality Controversy and Comparative Corporate Governance”, 47 Am. J. Comp. L. (1999) pp. 583 et sqq. (discussing the history of related ideas); Bayern, “Of Bitcoins”, supra, note 7 (citing sources).

55 Hansmann, Henry B., “The Role of the Nonprofit Enterprise”, 89 Yale L.J. (1980) pp. 835 et sqq., at p. 838 (“A nonprofit organization is, in essence, an organization that is barred from distributing its net earnings, if any, to individuals who exercise control over it, such as members, officers, directors, or trustees.”).

56 Cf. Bayern, “Of Bitcoins”,supra, note 7, at p. 1495.

57 For a recent discussion of veil piercing, see Macey, Jonathan and Mitts, Joshua, “Finding Order in the Morass: The Three Real Justifications for Piercing The Corporate Veil”, 100 Cornell L. Rev. (2014) pp. 99 et sqq..

58 See Smith, Douglas G., “A Federalism–Based Rationale for Limited Liability”, 60 Ala. L. Rev. (2009) pp. 649 et sqq., at n. 12 (collecting recent sources).

59 See O’Neal, Forrest Hodge and Thompson, Robert, O’Neal's Close Corporations, 3rd ed.(1997), § 1.10, at pp. 48.

60 See Rev. Unif. Unincorporated Nonprofit Assoc. Act § 8 (2008) cmt. (“Courts have pierced the corporate veil of nonprofit corporations… . The fact that members of nonprofit corporations for the most part do not have an expectation of financial gain, as compared to shareholders of a for profit corporation, should mean that there will be fewer types of cases than those involving for profit corporations where the veil piercing doctrine will be held to be applicable to nonprofit corporations.”).

61 E.g., Singer, Peter, Practical Ethics 2d ed. (Cambridge: Cambridge University Press, 1993) at pp. 182–90 (arguing for a limited conception of rights for severely disabled humans).

62 Cf. Solum, “Legal Personhood”, supra, note 45. 63 See Rob Atkinson, “The Low Road to Cy Pres Reform: Principled Practice to Remove Dead Hand Control of Charitable Assets”,

58 Case W. Res. (2007) pp. 97 et sqq. (discussing the problem of dead–hand control, primarily in the context of not–for–profit organizations).

64 See Bayern, Closely Held Organizations, supra, note 5, at pp. 273–83 (discussing deadlock in LLCs and the possibility of judicial dissolution).

65 See RULLCA § 701(4)–(5).

66 See RULLCA § 705.

67 RULLA § 705(a)(1). 68 See supra, note 4.

* Larry & Joyce Beltz Professor, Florida State University College of Law. BS, Computer Science, Yale University; JD, University of California, Berkeley School of Law. For helpful discussions related to the ideas in this article, I thank Rob Atkinson, Thomas Burri, Mel Eisenberg, Mark Gergen, Jay Kesten, Mark Spottswood, and Don Weidner. I also thank Thomas Burri and Isabelle Wildhaber for convening the conference at the University of St. Gallen that inspired this paper. A version of this paper was originally published online in the United States in the Stanford Technology Law Review; it appears here with permission of that journal.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

European Journal of Risk Regulation
  • ISSN: 1867-299X
  • EISSN: 2190-8249
  • URL: /core/journals/european-journal-of-risk-regulation
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed