Robotics and the law

Citation

Calo, R. (2013), "Robotics and the law", Industrial Robot, Vol. 40 No. 2. https://doi.org/10.1108/ir.2013.04940baa.001

Robotics and the law

Article Type: Viewpoint From: Industrial Robot: An International Journal, Volume 40, Issue 2

 

The technical challenges to deliver on the transformative potential of robotics are considerable. Robot ethics largely assumes these challenges will be overcome, at which point the need for serious thinking about the impact of robotics on human life will be even more pressing.

Robotics and the law makes no such assumption. Its interest, as I conceive of the field, is in developing the best policy and infrastructure to receive robotics. And among its central concerns is that we may never realize robots’ potential for reasons having little to do with the state or cost of the technology.

There are several ways the law may interfere with the development of robotics. The first sounds in intellectual property. Robotics requires collaboration. Companies will not be in a position to develop every robot from scratch; they will need to build on one another’s work. At the same time, individual firms will need monetary incentives to innovate.

Software and hardware companies faced a similar challenge in the context of personal computers and now mobile computing. Things are not shaping up well by many accounts: companies make defensive and offensive use of their patents in court and even amass intellectual property “war chests” that tax the industry and create barriers to entry. Robotics has the opportunity to learn from these mistakes.

A second legal challenge is privacy. Practically by definition, robots are equipped with the ability to sense, process, and record the world around them. One of the principle uses to which we have put robotics has been surveillance. The prospect of domestic drones in particular has occasioned massive public pushback (Calo, 2011b). Unwisely, in my view, the Federal Aviation Administration has refused to make adequate privacy practices a condition of receiving a license to operate drones in the USA.

But drones are only the beginning. Robots capable of recording and transmitting information – machines whose very anthropomorphic design have the potential to interrupt solitude – are beginning to enter our work and living spaces (Calo, 2010, 2012). Absent adequate safeguards for privacy and security, citizens and consumers may well reject this technology wholesale.

Yet another challenge is liability. Arguably the best model for personal and service robots is to design platforms that permit for third party contribution (Calo, 2011a). Computers and phones have apps – robots could too. Instead of buying a robot for every task, individuals and business could buy robots with no dedicated purpose and download or even write task-specific software. Just as Apple need not write every app for its iPhone, iRobot need not come with every use for its new telepresence robot AVA.

The downside to “open robotics,” however, is that liability is not obvious. If something goes wrong, is the platform liable, the app developer, or the user? Courts domesticated liability for software through aggressive enforcement of warranties and other, artificial limits on recover. With physical damage on the line, however, it seems unlikely courts will be able to avail themselves of the same expedients in the context of robotics (Calo, 2011b).

A final issue (for purposes of this Viewpoint) has to do with regulatory hurdles. Approval of, for instance, robotic medical devices is lengthy, involved, and uncertain. The problem is partly that the Food and Drug Administration has few analogs for many robotic systems that permit it to streamline the application process. As a consequence, roboticists are developing and marketing their products abroad or even abandoning the medical field in search of less onerous regulatory contexts. Some robots with the potential to bring comfort and improve human health never see the light of day.

These and similar challenges are hardly the exclusive province of lawyers. Other disciplines from engineering to business are in a position to mitigate the harms and provide solutions. But no roadmap to a transformative robotics industry is complete without reference to law and policy. Good laws remove barriers to innovation while preserving incentives for safety, privacy, and other values. Bad laws lessen the chances we will even face to the valid concerns of robot ethics.

Ryan CaloBased at University of Washington School of Law, Seattle, Washington, USA