Building Robots Without Ever Having to Say You’re Sorry

The European Parliament’s draft recommendations for governing the creation and use of robots and artificial intelligence includes rights for “smart” robots

3 min read
Photo of robot face
Photo: Getty Images

In January, the Legal Affairs Committee of the European Parliament put forward a draft report urging the creation and adoption of EU-wide rules to corral the myriad issues arising from the widespread use of robots and AI—a development, it says, is “poised to unleash a new industrial revolution.”

It’s an interesting read, and a valiant effort to get a handle on how to standardize and regulate the ever-expanding robot universe: drones, industrial robots, care robots, medical robots, entertainment robots, robots in farming—you name it, they’re all in there.

Beginning with Frankenstein’s monster, Prague’s golem, and Karel Čapek’s robot and ending with a code of ethics for robotics engineers and some daunting lists of “shoulds” for robot designers and end users, the 22-page worry catalog toggles between practical concerns about liability, accountability, and safety—who’s going to pay when a robot or a self-driving car has an accident?—and far-ranging ones about when robots will need to be designated “electronic persons,” and how we will ensure that their creators make them good ones.

The practical concerns addressed include a call for the creation of a European agency for robotics and artificial intelligence to support the European Commission in its regulation- and legislation-making efforts. Definitions and classifications of robots and smart robots need to be detailed, and a robot registration system described. Interoperability and access to code and intellectual property rights are addressed. Even the impact of robotics on the workforce and the economy are flagged for oversight.

The “electronic persons” discussion, tucked halfway through the report, caught everyone’s attention—perhaps because it’s much more fun to catastrophize about HAL 9000 and Skynet than it is to ponder robot insurance requirements. And because personhood—what it legally means to be recognized as a person—is such a loaded topic.

Mady Delvaux, a Luxembourg member of the EP and the report’s author, attempted to clarify the designation of what a limited “electronic personality” would be, saying that it would be comparable to the standing that corporations have as legal persons, making it possible for them to conduct business, limit liability, and sue or be sued for damages.

But we haven’t finished addressing legal definitions of personhood for women, children, and higher-order animals like chimpanzees yet. Are we really ready to take on robot e-personhood?

I called Joanna Bryson, reader in the department of computer science at the University of Bath, in England, and a working member of the IEEE Ethically Aligned Design project, to ask her what she thought, having just read the Reddit Science “Ask Me Anything” she did about the future of AI and robotics. Her response? As soon as you put the word “person” in the draft, you’re probably in trouble.

She told me about Australian law professor S.M. Solaiman’s article “Legal Personality of Robots, Corporations, Idols and Chimpanzees: A Quest for Legitimacy,” which argues that corporations are legal persons but AIs and chimpanzees aren’t. Legal persons must know and be able to claim their rights: They must be able to assert themselves as members of a society, which is why nonhuman animals (and some incapacitated humans), and artifacts like AIs should not, according to Solaiman, be considered legal persons.

But then Bryson said something I had not considered. Since robots are owned—they are in a sense our machine slaves—we can choose not to build robots that would mind being owned. We aren’t obliged to build robots that we end up feeling obliged to, says Bryson. So instead of assuming that an ethically challenged future saturated with sentient machines is inevitable, we could choose to maintain agency over the machines we are building and defy the technological imperative. Could we do it? Or are we so in thrall to the notion of creating artificial life, monsters, and golems, that it’s irresistible?

This article appears in the March 2017 print issue as “Do We Have to Build Robots That Need Rights?”

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

This article is part of our special report on AI, “The Great AI Reckoning.

"I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓ Show less