Race, Necropolitics, and the Human in an Age of Intelligent Machines
This post is by Sylvester A. Johnson, Professor and Director of the Center for Humanities at Virginia Tech.
In July of 2016, the Dallas Police Department set a precedent when officers deployed a Remotec robot to detonate one pound of C4 explosive in order to kill an African American suspect, Micah Johnson, who was suspected in the shooting deaths of five police officers. When news first broke that a robot designed for warfare against military enemies abroad had been deployed to kill a civilian shooting suspect, chiefs of police throughout the country immediately affirmed the killing and noted they possessed similar robots and would have taken the same action. For decades, science fiction films and books have conjured futuristic scenarios of robots and other intelligent machines killing humans or running amok. So, this killing in 2016 came as a shock to many casual observers accustomed to associating intelligent machines with fiction. For those who have been following the dizzying pace of developing AI robots and machines’ intelligence, however, things are less surprising. For everyone, the fact that intelligent machines have now begun to play a role in matters of life and death opens a new chapter in the history of relations between humans and technology. It exposes the fragility of common assumptions about technology, race, and agency. And it implies that the domain of technology and necropolitics — administering death and brutality — will have to be theorized anew in an age of intelligent machines.[1]
We are indebted to Achille Mbembe for theorizing what he has named necropolitics, the vertical assemblages of power affecting the “generalized instrumentalization of human existence and the material destruction of human bodies and populations.” Moving beyond Michel Foucault’s theorization of biopolitics — along with Agamben’s analysis of bare life and Carl Schmitt’s theorization of the political — Mbembe urges that we interpret sovereignty and the nomos (the basis for “right”) of our contemporary world through the lens of social data concerning the technologies of death, and political will to destroy populations of highly vulnerable people.[2]
The robotic killing of Micah Johnson echoes the historical and enduring pattern of racial disparity in policing. Experts have repeatedly demonstrated that this pattern is a continuing form of lethal violence that US police departments use to engage Black civilians. (Consider, by contrast, that municipal police pursuing the White shooting suspect Dylan Roof swaddled him in a bullet-proof vest for protection after Roof murdered en masse a group of Black parishioners in South Carolina.)[3] Johnson, furthermore, was by no means the first suspect to create a violent stand-off with municipal police, but he will now stand in memory as the first US citizen on domestic soil to be targeted and killed by means of a state-deployed robot.[4]
Why should we query the relationship between Artificial Intelligence (AI) and necropolitics? Succinctly, we live at a time when the furthest aspirations of military research in AI and human enhancement (synthetic biology and human-machine hybrids) are transforming the ability to alter our humanity and surveil target populations. What once seemed an unlikely convergence between questions of human identity and national security has now become imminent. In this context, the so-called new materiality — to which scholars of religion have become attuned — gains new urgency and prompts important rejoinders. The provenance of the most far-reaching and impactful AI machines and human enhancement technology is militarism, particularly understood as that greater complex of military-industrialism that links up elite levels of military executives, defense contractors, research universities, global commercialism (particularly weapons sales), technological corporations, and private banks. The race to achieve human-level intelligence in weaponized AI has emerged as a global pursuit, in tandem with plans for highly engineered biological human soldiers who might be combined with machines using implants, exoskeletons, and even methods of synthetic biology.
Early results of these ambitions include brain-computer-interface (BCI) architectures. One application joins a human to a machine-learning system. This system then monitors a soldier’s brain, signaling to accelerate human analysis of surveillance video feeds. Other BCI applications restore motor and cognitive capacity to injured soldiers using brain chip implants. Autonomous weapons systems already exhibit capacities far greater than that of the Remotec used to kill Micah Johnson. Lockheed Martin’s AI-enabled joint air-to-ground missile (JAG-M) features “fire and forget” technology. Once a fighter pilot launches the missile, the JAG-M missile does the rest, allowing the pilot to ‘forget about it’ and move on to another task while the missile interprets its video feed, distinguishes its target from the larger environment, plans its own course and angle of attack, and changes its navigation course on the fly to neutralize the target’s evasion tactics.
It is currently considered unconstitutional in the United States for weaponized AI to kill human targets without a human participating in making the decision to kill (a human “in the loop”). This should not elide the fact, however, that multiple autonomous weapons systems already exist with this capability. So, the technology is ahead of the policy. And it would be irresponsible to assume that the continuing development of autonomous weapons will be tamed by the current version of our legal theory.
This is complicated by the fact that national security paradigms have thrived on the perception of racial threat to reinscribe militarism as an expansive canopy traversing both domestic and foreign domains. The internment of Japanese Americans; the deployment of the FBI’s COINTELPRO against Black liberation activists, the Chicano movement, and the American Indian movement; and the counter-terrorism measures that target Muslims as an inherent threat to the nation and the racial civilization of the West — all evidence that state enemies are perceived beyond and within national borders.[5] This is why civilian policing has become increasingly integral to counter-terrorism operations. As military research yields more advanced forms of weaponized AI and human enhancement, the repercussions for racially targeted domestic subjects appear ominous. This is true especially in light of the militarized policing that national security practices have normalized since the 1960s. Such was, after all, the point behind the Dallas Police Department’s decision — along with that of other police departments throughout the nation — to acquire military robots for civilian policing. The increasing stockpile of weaponized AI by police departments is not for foreign enemies but for domestic targets.
There is yet another dimension to this subject that will demand critical attention as well: the humanization of machines. Humans are rapidly accelerating the pace at which intelligent machines become more capable of performing cognitive functions — thinking and reasoning. Deep Mind’s AlphaGo algorithm is a widely noted example, producing unprecedented insights into what is perhaps the world’s most complicated board game and easily surpassing the most skilled players. IBM’s Project Debater is another, able to listen and respond in real-time to humans debating a topic to provide a richly data driven and shrewdly formed counter-argument on a given topic. Given the influential philosophical tradition that asserts thinking and reasoning are central to human identity, the advance of intelligent machines is raising novel concerns about the relationship between people and intelligent machines. At the heart of these questions are the debates over what it means to be human, whether machines can become human, and to what extent this will intersect with established practices of racialization.[6]
Various writers, from Sylvia Wynter to Ray Kurzweil, have examined the category of the human in ways that interface with the implications of AI. Wynter has critiqued the long history of Europeans developing a racial conception of the human being by defining White men as the ideal human type, and all others as less than the ideal — i.e., less human or not human. She argues that the European tradition of interpreting biological science has been a preeminently racist enterprise, which promotes White supremacism by locating evidence for race in the body. Wynter has examined neuroscience in order to explain how sociogenic factors (social reality, intersubjectivity, etc.) can change the way bodies actually work. And she emphasizes that sociogenic (i.e., socially derived or constructed) and phylogenic (i.e., biological) factors are equally central to constituting humans.[7] Wynter’s intervention should lead us to develop a more critical approach to the “human” in the “humanization” of machines. As the human category is burdened by racial and colonial freight (the category has frequently functioned to distinguish among races, those fully human versus others), we could only naively regard the humanization of machines as a raceless enterprise.
It is precisely on this score that Kurzweil’s assessment of human ontology becomes so useful. Kurzweil is completely uncritical of the racial architecture of science and politics. In fact, he has been rightly critiqued for his utopian view of technology: he assumes, for instance, that structurally-engineered racial disparity and necropolitics will play no role in the future of technology. His robust engagement with an informational paradigm for interpreting human ontology, however, is productively wed to the critique of race that Wynter advances. That is, by emphasizing informational capacity as central to not only humans, but also machines — and more expansively, the material entities of the universe — the interventions of Kurzweil and Wynter together enable an approach to assessing the human-object binary that contravenes the Platonic inheritances of Western epistemology; material no longer seems so foreign to informational capacity. Rather, it emerges as the means of that capacity.[8] If informational capacity — the ability to embody or contain, relay, and manipulate information to produce knowledge, thinking and reasoning–is foundational for what we have called human experience, then it becomes firmly debatable that informational machines are perhaps already becoming human.
The rapid ascent of intelligent machine systems witnessed over the past decade is surely a harbinger of things to come. It is most likely that we will meet with a continuing wave of unprecedented advances in technology capacity. The military-industrialism that continues to spawn the most ambitious AI applications and human-machine hybridization will not only produce socially beneficial technologies, but will also deepen the necropolitics that has shaped race governance. This, in tandem with the racial politics guiding the imagination of the human template for humanizing machines, presents a daunting challenge to those who might wish to instantiate a different politics for our human future, one not rooted in instrumentalizing disparity, death, and destruction of vulnerable populations. This will be the grand challenge for the race-critical guidance of technology in the years to come.
[1] Kevin Sullivan, Tom Jackman and Brian Fung, “Dallas Police Used a Robot to Kill. What Does That Mean for the Future of Police Robots?” The Washington Post, July 21, 2016. https://www.washingtonpost.com/national/dallas-police-used-a-robot-to-kill-what-does-that-mean-for-the-future-of-police-robots/2016/07/20/32ee114e-4a84-11e6-bdb9-701687974517_story.html, accessed 9/14/2016.
[2] Achille Mbembe, “Necropolitics,” Public Culture 15, no. 1 (2003): 11–40. Mbembe further theorizes the architecture or race in the digital future of necropolitics in the opening to his Critique de la raison negre (Paris: La Découverte, 2013).
[3] Dylan roof was eventually convicted of murder in December 2016 and was sentenced to death. https://www.theguardian.com/us-news/2017/jan/10/dylann-roof-sentenced-to-death-charleston-church-shooting.
[4] Racialized practices of policing and incarceration have been repeatedly documented by expert studies since the 1800s (Ida B. Wells Barnett produced multiple exposés of such). The perduring practice of state racism in law enforcement is a major factor for understanding the necropolitical implications of using AI designed for warfare in context of domestic policing. Representative studies include Khalil Muhammad, The Condemnation of Blackness: Race, Crime, and the Making of Modern Urban America (Cambridge: Harvard University Press, 2010); Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness (New York: New Press, 2012); Paul Butler, Chokehold: Policing Black Men (New York: New Press, 2017); and Andrea Ritchie, Invisible No More: Police Violence Against Black Women and Women of Color (Boston: Beacon Press, 2017). In 2011, the US military used a drone to kill Anwar al-Awlaki, a US citizen residing in Yemen at the time. Al-Awlaki became the first US citizen to be targeted and killed in an “extra-judicial” assassination, as he was classified as a terrorist and thus denied due process. See H. Jefferson Powell, Targeting Americans: The Constitutionality of the U.S. Drone War (New York: Oxford University Press, 2016).
[5] Karen Alonso, Korematsu v. United States : Japanese-American Internment Camps (Springfield, IL: Enslow, 1998). Ward Churchill Jim Vander Wal, The COINTELPRO papers : Documents from the FBI’s Secret Wars against Dissent in the United States (Cambridge: South End Press, 2002).
[6] Fei-Yu Wang et al, “Where Does AlphaGo Go: From Church-Turing Thesis to AlphaGo Thesis and Beyond” IEEE/CAA Journal of Automatica Sinica 3, no. 2 (April 2016): 113–20.
[7] Sylvia Wynter, “Unsettling the Coloniality of Being/Power/Truth/Freedom: Towards the Human, Afer Man, Its Overrepresentation — An Argument,” CR: New Centennial Review 3, no. 3 (2003): 257–37.
[8] Ray Kurzweil, The Singularity is Near: When Humans Transcend Biology (New York: Viking, 2005).
