Breaking the First Law

Asimov and killer robots in the academy.  

by Colin Broadmoor

On July 8, 2016, the Dallas Police Department killed Micah Xavier Johnson by intentionally detonating a block of C-4 a few feet away from his body. True to the perverse irony that characterizes so much of American policing, this improvised explosive device was delivered by a bomb disposal robot. In these dystopian days, it’s commonplace to see police actively using “less than lethal” technologies to kill, maim, and terrify, but, at the time, the method of Johnson’s death was notable enough to justify a New York Times article on the subject. Micah Xavier Johnson was the first—but almost certainly not the last—person to be killed by an American police robot.

Assassination-by-robot was neither inevitable nor a natural consequence of technological advancement, but it was predictable. In 1942, almost 75 years before Johnson’s death, Isaac Asimov released the short story “Runaround” that included the first iteration of his Three Laws of Robotics:

• First Law — A robot may not injure a human being or, through inaction, allow a human being to come to harm.
• Second Law — A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
• Third Law — A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Later, in “The Evitable Conflict,” Asimov formulated a “Zeroth Law”: A robot may not harm humanity, or, by inaction, allow humanity to come to harm. In 1950, these and other stories exploring the ethics and limitations of robotics were re-released in the collection I, Robot.

Assassination-by-robot was neither inevitable nor a natural consequence of technological advancement, but it was predictable.

In 2020, Asimov’s Laws of Robotics read like an alternative history of a world in which a press release celebrating heavily armed swarm-drones would be seen as abhorrent rather than the norm. From any rational or moral perspective, a world without killer robots is superior to a world in which multinational corporations sell robots specifically designed to kill and control. Most people would prefer to live without the possibility of being murdered by a machine; however, “most people” have no voice in the way deadly technology is produced and deployed across the world. Instead, the power to make these life or death decisions is held by politicians who love war and fear protests, along with arms dealers who enrich themselves by supplying these weapons of repression and imperialism.

Thanks to the fame of I,Robot and the later Will Smith film of the same name, people beyond the fringes of sci-fi fandom likely have at least a passing familiarity with the First Law. Asimov’s ideas pose a problem for those individuals and organizations that rely on fatal technologies to exercise power: in a consumer society, how do you neutralize a popular, easily grasped idea disseminated through media? Their answer is to use the apparatus of the academy to police how we understand the radical potential inherent in art and, especially, in science fiction.

Sci-fi is powerful because it offers alternatives to the present moment by imagining different worlds—the first step to changing the world we live in. Writers like Asimov, Octavia Butler, Ursula K. Le Guin, Kim Stanley Robinson and so many others have used science fiction as a medium to challenge the unexamined assumptions that undergird systems of social injustice. In the past, academics could safely dismiss sci-fi as “genre fiction” or “pop culture” unworthy of study or serious consideration. But science fiction’s ability to push at the edges of human imagination makes it perfect material for Debord’s spectacle. Sci-fi is valuable but dangerous. The success of sci-fi media franchises, especially in television and cinema, makes the genre too influential to be ignored any longer. In recent decades, the academy has shifted from deprecating the artistic merits of sci-fi to gatekeeping its interpretation. Scholarly treatments of Asimov’s laws offer a particularly useful example of the process by which academia delegitimizes sci-fi’s power of radical prophecy because the contradictions and conflicts of interest are so egregious. Asimov’s First Law is so fundamentally sensible that it takes a PhD to realize why it’s “wrong.” Or maybe not a PhD—maybe just an endowment from an arms manufacturer.

In the past, academics could safely dismiss sci-fi as “genre fiction” or “pop culture” unworthy of study or serious consideration. The success of sci-fi media franchises makes the genre too influential to be ignored any longer.

In “Beyond Asimov: The Three Laws of Responsible Robotics,” Robin R. Murphy (Texas A&M University) and David D. Woods (Ohio State University) lament, “The three laws have been so successfully inculcated into the public consciousness through entertainment that they now appear to shape society’s expectations about how robots should act around humans.” Besides science fiction scandalously influencing the way society thinks about safe technology and public health, we might wonder what is so bad about setting an expectation that a technology should not harm or kill people?

Murphy and Woods have little time for such concerns:

[The First Law] is already an anachronism given the military’s weaponization of robots, and discussions are now shifting to the question of whether weaponized robots can be “humane”… Even if current events hadn’t made the law irrelevant, it’s moot because robots cannot infallibly recognize humans, perceive their intent, or reliably interpret contextualized scenes.

Calling the belief that a technology should not harm humans “anachronistic” and “irrelevant” seems excessive. After all, we don’t say the Hippocratic oath is irrelevant because of crimes against humanity performed under the guise of medicine such as those perpetrated by the Nazis or the U.S. Tuskegee “syphilis studies.”

The argument by Murphy and Woods forecloses on the possibility Asimov created of a future where robots do not kill humans. They present killer robots as a fait accompli: robots are created to kill humans now; therefore, robots will always be created to kill humans. This thinking would be abhorrent if applied to any other social ill, but perhaps our academics do not see killer robots as a social ill.

Dr. Murphy is the “Raytheon Professor of Computer Science and Engineering at Texas A&M,” and the article notes tell us Dr. Woods’s perspective “was based on participation in the Advanced Decision Architectures Collaborative Technology Alliance, sponsored by the US Army Research Laboratory.” The disconnect between the weaponized reality of technology and their bloodless academic rhetoric of “responsible robotics” is telling given their proximity to the war machine.

Murphy and Woods present killer robots as a fait accompli: robots are created to kill humans now; therefore, robots will always be created to kill humans.

On a recent product page for The StormBreaker®, Raytheon boasts:

The StormBreaker® smart weapon gives operators an upper hand in combat by hitting moving targets in some of the worst weather conditions. The winged munition autonomously detects and classifies moving targets in poor visibility situations caused by darkness, bad weather, smoke or dust kicked up by helicopters. Poor weather and battlefield obscurants continue to endanger warfighters as adversaries rely on these conditions to escape attacks.

These autonomous machines cannot, according to Drs. Murphy and Woods, “infallibly recognize humans, perceive their intent, or reliably interpret contextualized scenes” well enough to prevent them from killing people, but according to Raytheon, they can “detect and classif[y] moving targets in poor visibility situations” well enough to blow them to pieces. In a rational world, the admission that machines cannot consistently distinguish between humans or human intentions would be a reason to ban machines that kill, but in the world we live in—the world in which “Raytheon professors” exist—this instead becomes an excuse to jettison the prohibition against killer robots…an inconvenient prohibition if you are one of the world’s largest arms dealers.

The two academics use their knowledge of death machines to provide their own improved First Law for truly responsible robotics. Instead of the hopelessly irrelevant “A robot may not injure a human being or, through inaction, allow a human being to come to harm,” Murphy and Woods suggest “A human may not deploy a robot without the human–robot work system meeting the highest legal and professional standards of safety and ethics.”

The possibilities for a non-militarized robotics expressed by Asimov are too dangerous to be taken seriously by a world in which killer technology is big business.

This is wonkishness at its most virulent. Murphy and Woods replace Asimov’s clear-cut prohibition on robots built to harm humans with a meaningless gesture toward the regulatory power of the professional managerial class—a class already implicated in the development and distribution of fatal technologies. This rendering of Asimov’s axiomatic “a robot may not injure a human” into toothless HR waffle  to serve corporate interests was itself predicted in sci-fi media such as Alien and Robocop, but here we see the intellectual sleight of hand that justifies robotic drones annihilating wedding processions.

The possibilities for a non-militarized robotics expressed by Asimov are too dangerous to be taken seriously by a world in which killer technology is big business. It falls to the academy to “interpret” visionary art in a way that renders it unthreatening to power. Asimov’s First Law of Robotics has already been broken, but we can imagine and work toward a world in which it becomes a reality. Unfortunately, our chances of achieving that goal are jeopardized by the fact that we are now on course to violate the Zeroth Law: A robot may not harm humanity. The US Air Force is already contracting the war machine to create an “R2D2-like” AI that will be able “to take hold” of fighter jet controls. The project’s name, Skyborg, eerily reminiscent of both Skynet and the Borg, is probably nothing to worry about.

We have limited time left to create a sustainable and just future. Sci-fi will be a vital tool in that effort because of its ability to capture our social imagination. Visionaries, artists, and dreamers stand on one side in this struggle and arms dealers and cultural gatekeepers on the other. We must ask ourselves which of these visions for the future is truly fictional: the one that says we can live in a world where robots are not built to kill, or the one that insists our legions of death machines will remain safely controlled by the most vigorous industry standards?

---