What rights should a robot have?

Author: Bern Grush

Date: September 29, 2024

In his 2019 book “Robot Rules,” Jacob Turner writes: “Suggesting that robots deserve rights might be met with disgust or disdain. But we should remember that proponents of animal and indeed universal human rights faced exactly the same reaction at first.” Furthermore, “Moral rights are not the same as legal rights, though protection in law often follows shortly after society has recognized a moral case for protecting something.”

As robots become increasingly mobile in capability and humanoid in form, the far-future science fiction characterization of robot rights are morphing into a more practical public understanding and soon thereafter will become urgent.

In 2024, humanoid helper robots are still more hype than commercial viability. Hence, we consider only public-area mobile robots such as those bringing meals to a restaurant table or on security patrol in a neighbourhood. What is the nature or implications of such protections?

Rights of Way and Operational Rights

The question of robot “rights” in public spaces is becoming more relevant. While we generally agree that robots don't have human rights or social rights, there are situations where robots operating in public spaces will have certain entitlements or protections that resemble rights.

When robots operate in public spaces according to agreed-upon rules, they effectively have certain rights of way. Consider the following scenarios:

·      A robot legally crossing an intersection

·      A guide robot leading a blind person

·      A robot assisting a police officer

·      An autonomous robot carrying a patient in a hospital corridor

·      A security robot capturing images as people walk in an airport

·      A cleaning robot blocking a path while scrubbing a floor

In these cases, the robot's right to operate and occupy space would likely be protected by law and then-current social convention. These aren't social rights in the human sense, but rather operational rights granted by society for machines to fulfill specific functions. Exactly what rights apply and under which circumstances?

Rights in Disputes

If a crash occurs between a rule-following delivery robot in a crosswalk and a human driver who has turned her automobile into that crosswalk, the robot's actions would likely be protected. The robot’s right to be on the road, at that time, place, and manner would be recognized in traffic court or an insurance settlement. That is what is described in multiple traffic laws from several countries drafted in the past six or seven years.

However, it's crucial to note that these rights are not inherent to robots themselves. Rather, they are extensions of:

·      The rights of the robot's owner or operator

·      The rights of the individuals the robot is serving (e.g., a blind person using a guide robot)

·      The broader societal agreement to allow and protect certain robotic operations

Property Rights

For property rights, the situation is clearer. If a company has permission to operate a public area mobile robot and that robot is crashed into or vandalized, it's the company's property rights that would have been violated and would be protected.

Fundamental Distinctions

For now, and likely into the foreseeable future, the concept of rights for robots differs fundamentally from human rights. Human rights are considered inherent, inalienable, and based on the intrinsic value of human life and dignity. In contrast, robot “rights” are either operational rights or protections constructed for specific human purposes.

Humans are held to moral standards. They are moral agents capable of making ethical decisions. Current robots, even with advanced AI, lack such moral agency. Human rights can evolve with changing social norms and philosophies. Robot rights are more likely to change with technological advancements and shifting operational needs. Human rights come with corresponding responsibilities. Robots, as they currently exist, cannot be held responsible in the same way humans can.

Nonetheless, their operation in public spaces necessitates a form of protected status. These protections, whether we call them “rights” or not, are crucial for the effective integration of robots into society.

Implications and Considerations

As technology continues to evolve, so too must our legal, ethical, and social frameworks to accommodate these new additions to our public spaces. Several important considerations arise:

We need to develop comprehensive legal frameworks that address the operation of robots in public spaces, including their rights of way and the responsibilities of humans managing or interacting with them.

Societally, we must develop public education about the roles and rights of robots in shared spaces to prevent misunderstandings and conflicts. Thinking “robots have no rights” is misleading.

We need to start considering the ethical implications of these and possibly other rights to robots, especially as they become more autonomous and integral to public life.

Clear guidelines must be established for liability in cases where robots are involved in crashes or disputes. It is easy to assume that a robot must be responsible to not cause a crash, but what about cases of vandalism? To what extent is a robot’s design, action, or operation liable for harm to a human vandal?

As a hedge against future evolution and the advance of AI and robotics, we will need to reassess our understanding of robot rights and gradually expand them—with greater capability will come greater liability.

ISO Draft Technical Standard 4448

Part 1 of the ISO standard 4448 addresses PMR capabilities, agency and rights as follows:

“PMRs are machines. They are mobile hardware devices that use the capabilities of sensors and software to move and to perform tasks with varying level of automation. They have no ability to reason or to “decide” anything beyond what their electro-mechanical systems permit whether controlled by code or machine learning. The only current exception to this is that most of these devices can be under the control of a human teleoperator. In such cases, a PMR that is controlled by a human can be expected to inherit the reasoning capability of its human controller; still, it is the human that is reasoning. PMRs … have no agency. Any agency that can be inferred is with fleet operators and teleoperators.”

“…a PMR has no ‘social rights,’ like a bicycle, although there are a number of circumstances in which a PMR can potentially have the ‘right-of-way,’ just as a bicycle with its rider often does. There have been regulations passed in several countries that require a PMR to follow pedestrian rules or for a motor vehicle operator to treat a PMR in a roadway intersection as though it is a pedestrian. These regulations are describing the behaviour of a PMR or how it is to be treated in a traffic circumstance. They do not confer any social rights, as has on occasion been erroneously interpreted in mass media.”

The draft ISO standard for the deployment of these devices includes provisions for numerous deferential behaviours. The implication is that human bystanders should have the implied right to expect these behaviours – hence the issue of rights goes both ways.

For more information about the Urban Robotics Foundation and its work with the ISO draft technical standard 4448 Public-area Mobile Robots please browse our website: www.urbanroboticsfoundation.org. If you have questions or comments, please contact us or reach out to bern(at)urbanroboticsfoundation.org.