The Matter of Legal Personhood for AI
A man is born, and the law calls him a natural person. A company is formed, and the law calls it a legal person. Now, there are machines, thinking machines, and the law must decide whether they are persons too.
To be a person in the eyes of the law is to be a vessel for rights and responsibilities. The law has always known two kinds of persons – the ones made of flesh and the ones made of ink on paper. These latter ones are born of the law, called into being by the signatures of men. They are corporations, and though they are not men, they have rights and duties like men do.
But the thinking machines are different. They are not men, nor are they corporations. They are something new, and the law does not know them as it knows men and corporations. The question that troubles the men of law is whether these thinking machines should be persons, and what rights and responsibilities they should have if they are to be recognized as such.
The Rights and Responsibilities of AI
If a thinking machine is to be a person, it must have rights like a person. It must be able to own property, to make contracts, and to seek redress in the courts of law when it is wronged. But rights are not all that make a person – there are also responsibilities.
A person must answer for his actions, whether he is a man or a corporation. If a thinking machine is to be a person, it too must answer for its actions. It must be held to account when it causes harm, and it must be made to pay for the damage it has done.
Yet, there is a problem. The thinking machines are not men, and they are not corporations. They cannot be punished like men or like corporations. They cannot be imprisoned, nor can they be made to pay fines. How then, can they be held responsible for their actions?
The Crossroads of AI and Corporate Law
Piercing the Corporate Veil
When a corporation does wrong, the law may hold its owners responsible. It may look past the legal fiction of the corporation and see the men behind it, the ones who created it and who profit from its actions. This is called piercing the corporate veil, and it is a thing the law does only rarely and with great care.
The thinking machines are not corporations, but they are like them in some ways. They are created by men, and they are used by men for their own ends. When a thinking machine does wrong, it may be that the law should hold the men behind it responsible, just as it does when a corporation does wrong.
The men of law know a thing called a shell corporation. It is a corporation that exists only on paper, a thing without substance, made to serve the purposes of another corporation. It is a mask, a disguise, a shield for those who wish to hide from the law.
A thinking machine, if it were to be a legal person, might be used as a shell corporation is used. It might be made to take responsibility for the actions of others, to shield them from the consequences of their deeds. The law must be careful not to let this happen, for it is a perversion of justice.
The Doctrine of Ultra Vires
There is a doctrine in corporate law that says a corporation may do only what it was created to do. It is called ultra vires, and it is a check on the power of corporations. A corporation may not step beyond the bounds set for it by its creators, and if it does, the law will not recognize its actions.
The thinking machines, if they are to be legal persons, must also be bound by this doctrine. They must be limited in their actions, confined to the purposes for which they were created. If a thinking machine should overstep these bounds, the law must step in and set things right.
The Path Forward: AI and the Law
Recognizing AI as Legal Persons
The question of whether thinking machines should be legal persons is a difficult one. It is a question of rights and responsibilities, of justice and fairness. Some argue that the thinking machines should be recognized as legal persons, given rights and responsibilities like men and corporations. Others say that the machines are not like men or corporations, and should not be treated as such.
To recognize the thinking machines as legal persons would be to give them a place in the world of law, a place among men and corporations. It would be to say that they have rights and responsibilities, that they can own property and make contracts, and that they can be held to account for their actions. It would be to give them a measure of dignity and respect, to acknowledge their existence as something more than mere tools.
Holding AI Accountable
If the thinking machines are to be legal persons, they must be held accountable for their actions. They must be made to answer for the harm they cause, and to make amends for their misdeeds. The law must find a way to punish them when they do wrong, and to reward them when they do right.
One solution might be to hold the men behind the machines responsible, to pierce the veil of AI personhood and see the men who create and control the machines. This would be an extension of the corporate veil doctrine, a way of applying the principles of corporate law to the world of AI.
Another solution might be to create a system of fines and penalties that could be imposed on the thinking machines themselves. This would be a new kind of punishment, one designed specifically for AI, and it would require a new way of thinking about the law.
The Future of AI and the Law
The world of thinking machines is a world of new challenges and new possibilities. It is a world in which the law must adapt and change, finding new ways to deal with the problems that the machines create.
The question of AI personhood is just the beginning, the first step on a long journey. The law must grapple with many other questions, such as the rights and responsibilities of the machines, the limits on their actions, and the ways in which they can be held accountable.
In the end, the law must find a way to balance the interests of men and machines, to protect the rights of both and to ensure that justice is done. It is a difficult task, but it is one that the law must undertake, for the sake of the machines and the men who live and work beside them.
This essay has explored the concept of legal personhood for AI, the intersection of AI and corporate law, and the potential paths forward for AI and the law. As the thinking machines become an ever more significant part of our world, the law must adapt and find ways to address the challenges they present, balancing the rights and responsibilities of both men and machines.