“[I]t is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.” [1]
Abraham Maslow
The story of the building of Babel in Genesis does not begin with a plan to build a city and a tower, but with a seeming innovation in brick-making (Gen 11:3). Having unlocked the new potential for construction of stronger bricks, the plan to build Babel and its tower follows. The technological innovation catalysed the vision and made it conceivable, rather than the vision driving the innovation.
The distinction between these two phases in a new technology—a specific technological innovation and the world created by its expansive implementation—is worth reflecting upon. When talking about the automobile, we might consider it from the perspective of the internal combustion engine that made it possible. However, we might also consider it from the perspective of the world that was reformed around it. By itself, the internal combustion engine was of limited use: the true potential of the technology could only be released as the whole world was reordered around it. For the man with a car, the world is never traced with enough asphalt.
Much of the modern world can be considered as a negotiation between the affordances of the automobile and various other goods. The Locomotives Act of 1865 required a 2mph speed limit in cities, towns, and villages and 4mph in the countryside and, most famously, that a man waving a red flag walk ahead of such vehicles to warn people and horses in their path. Such legislation, while clearly designed with heavy steam-powered vehicles in mind, illustrates some of the challenges that the car had to overcome to achieve widespread adoption. To accept the automobile, society would have to have much higher tolerance for road deaths; it would also need to remove most horses, pedestrians, and children from streets and highways.
The world of the car most obviously involves our vast infrastructure of roads. However, it also involves radical transformations in the form of our settlements. Villages and their local economies weakened while suburbia grew. In those places that threw their lot in most fully with the new technology, cities could be carved up into vast archipelagos of separated locations by the roads built through them. Locality, neighbourhood, and community were weakened, not least as the entangled threads of old town centers became unravelled. With the advent of the car, the uprooting of the population and the decentering of their communities were accelerated. People formerly settled in and known to each other in communities for generations were increasingly strangers to most of their neighbours in communities with a rapid circulation of inhabitants. Roads could cut former communities in half or reduce them to stagnant backwaters, bypassed by traffic, but now lacking the capacity to maintain a robust local economy.
With the weakening of locality and community, power, oversight, and provision of welfare have also migrated towards higher and centralized authorities and the weight of people’s self-understanding shifted to less rooted and given identities. The car transformed people’s relationship to the state and the police, radically increasing their exposure to and habituating them to government bureaucracy, surveillance, regulation, and law enforcement. Increased mobility changed the character of national borders and identities. It came with transformations in the form of the economy and business with the rise of scientific management and, with the reduction of the power of local community, the increased power of business and government in intermediating, managing, representing, directing, and regulating society. Its fragmentation of place and society and empowerment of the individual over against the community played its part in encouraging the rise of countries of detached consumers. Prevailing forms of church are in many respects also down the road from the car.
Making the world navigable to the car also came at a considerable environmental cost. Roads tore through the countryside, fracturing and degrading habitats. The car has created extreme new dependencies and risks. Questions of sustainability, pollution, and environmental impact are hard to avoid. The need for oil has also given rise to new dynamics in international politics, producing tensions, shifting power, and sparking wars.
The immense yet disruptive potential of the technology of the car required innumerable decisions and trade-offs, not least on account of the considerable new risks to life it introduced. The world of the car is a heavily regulated world. It is a world of rules of the road, driving tests, speed limits and cameras, seat belts, police sirens, parking tickets and motoring fines, license plates, mandatory vehicle safety tests and certificates, prison sentences for vehicular manslaughter, emissions standards, the visual cluttering of our settlements with road signage and marking, laws against jaywalking and regulation of street-crossing for pedestrians, standardization of highway construction, road taxes, insurance, safety requirements for car manufacturers, congestion charges, road tolls, etc., etc. Such regulation has made driving considerably safer and, consequently, made people much more prepared to accept the trade-offs it involves. About three times as many people (4,886) were killed on British roads in 1926 than were killed in 2024 (1,633), a statistic more remarkable when you consider that there are twenty times the number of cars on the road today.
The potential gains of a technology like the internal combustion engine are gains that largely come to those who are prepared to reorder society around them. And, at a certain point, even those of us who do not own cars have no choice but to conform ourselves in various ways to the less hospitable world that the car has made for them. Much of the power of a technology or technique will come from its capacity to render the world controllable, manipulable, representable, replicable, navigable, or measurable. However, this power invariably requires trade-offs, as, for the technology to be truly effective, the world must be represented in terms of it and conformed to it. This process is never lossless: much of the reality of the world must be discounted, lopped off, fractured, or squeezed into an inhospitable framework. The technologies of money, for instance, provide immensely powerful and useful, yet limited representations of value, blinding us to much intangible, inalienable, non-fungible, and non-commensurable wealth, and to the distinction between the generation of the exchange value of money and the generation of true weal-th. To those seduced and enthralled by money, it can become the idol Mammon.
The “Machine” could be considered as the world that has been optimized for and conformed to technological affordances for the sake of maximization of wealth and power. It is a world in which our relationship with creation and each other is overwhelmingly intermediated by or substituted for by technology, our attention directed to and preoccupied with the machine, rather than to creation with the help of instruments (Hannah Arendt’s distinctions between the tool and the machine are important here). The dehumanizing effect of a society that has given itself over to the rule of technique is a common theme in technology criticism, prominent in the work of such as the Victorian social critic John Ruskin, for instance.
Jonathan Barlow summarizes some of these arguments in his essay that opens this Conversation. In his arguments against them, he focuses upon the theme of commensurability. He challenges scale-based thinking, which privileges agrarian life over city life, observing the prominence of the city in biblical eschatology. Barlow’s argument, however, seems to speak past the positions of many who advance the argument against the “Machine”: the point is not scale per se but the extent to which people are prepared to make Faustian bargains sacrificing their life together and the elevation of our humanity for technological powers.
As someone who travels a lot and spends much of my year in New York City, I have been struck by the sharp differences between the degree to which and the ways in which different cities and countries have negotiated these trade-offs. There are many cities that feel extremely human in many parts, while others feel more like anonymous technological, economic, and statist machines. It is also noticeable that many of the features that make many cities most livable and beautiful were only established or preserved because they preceded, were not subject to, or successfully resisted the demands of the car. The car was not rejected entirely, but humanity took dominion over the technology rather than thoughtlessly sacrificing valuable goods to it for the sake of its affordances.
Resistance to the Machine involves prioritizing proper human ends in regulating the movement from technological innovation to wide implementation, greatly privileging those ends in assessing the trade-offs for new technological powers and ordering the technology around the dominion that is proper to man. Resistance to the Machine also involves maintaining a firm purchase upon the central goods of humanity in ways that both reveal the narrowness of the technological frame and which, while using technology, refuse to give up the many inconveniencing goods that cannot be factored into or comported to it.
For all the talk of “AI alignment,” the development of AI has ironically revealed a radical lack of human alignment: competing AI companies with very different agendas, tensions between market demands and internal safety concerns, tensions between AI companies and the US government, and tensions between the US and China. Far from a careful, patient, and responsible development, regulation, distribution, and implementation of new AI capacities, we have witnessed a frenzied development process with mind-bogglingly immense sums of money plunged into infrastructure buildout, largely driven by people who are terrified of what might happen if some opposing company or country gets to super-advanced AI first and who are also terrified of the prospect of AI. The economic gamble upon AI is far too big to fail but may only really pay off if we comprehensively reorder our world around this new technology. We cannot afford not to “win,” but “winning” also comes with its own terrors, the rise of a new potential inhuman opponent with godlike powers. Pandora’s Box must be opened because not to do so would give less virtuous parties an immense advantage. This is a multipolar trap, not human dominion. The people developing AI do not come across as people who are in control.
The “world” envisioned by so many of the creators of AI, putting to one side the fact that many of them fear human subjugation or extinction by advanced AI, is not a very inviting world. It is a world of extreme and existential risks for humanity. It is an unmapped realm, unlike anything humanity has experienced before. It also functions as nothing less than an eschatology for vast swathes of the AI industry and its boosters. Even where it promises ease and plenty rather than wars with vast armies of drones and autonomous robots, reading between the lines one sees a world of precarious dependence upon government, AI companies, and unreliable AI agents, in whom power and wealth are concentrated, and where democracy likely withers as people become burdens to be managed rather than contributing citizens. It is a world where human labor is minimal, where human society is enervated, where human culture is marginal and likely atrophies, where literacy is diminished, and where we might be untethered from our bodies and our planet. It is a world of surveillance and control, a world mediated by the Machine. An anti-human eschatology is baked into so much of the technological development.
AI names numerous incredible technological innovations, with a myriad potential uses, many of them promising immense benefit to humanity. The problem with AI is not with these innovations narrowly considered. Rather, the problem is with the way that these building blocks are treated as irresistible invitations to construct a new Babel. Driven by a sort of ideological imperative towards the radical pursuit of technology on the one hand and a fatalistic understanding of technology’s inevitability on the other, the architects assure us that the Tower must be built and that—unless the Tower destroys us all—the future belongs to those who build it.
As Brad Littlejohn recognizes in his response, while we might argue for the incommensurability of human beings and AI in principle, in practice, the technology encourages us to treat human creation and AI output as interchangeable and AI output as substitutable for human creation, and it profoundly compromises our capacity to distinguish the two. Where there are not effective practical measures distinguishing AI output from human creations and preventing the substitution of AI for human creations in many areas, arguments about incommensurability in principle will count for little. Just as institutions like slavery blind us to the inherent dignity of human beings as creatures in God’s image, AI can blind us to the uniqueness of human beings; just as a genuine assertion of the principle of human dignity must threaten the institution of slavery, so it must oppose many uses of AI. Concerns about AI “slop,” for instance, do not arise from belief that AI generated output is commensurable in principle with human creations but with the fact that, in practice, it is increasingly masquerading as, substituted for, parasitic upon, competing with them, or destroying the realms and economies that sustain them. Bogumił Jarmulak’s superb response articulates something of what such resistance requires of us and what is at stake.
Sacrifice of human incommensurability in practice can unleash vast technological power. Resisting such sacrifices in a context where there are technologies offering power and ease to those who make them requires something near to heroic virtue on the part of individuals, or extensive and effective regulation on the part of societies and their institutions. The former, while valuable, is of little effect against the transformative ecological impact of such a new technology, and those taking it must increasingly operate in the wreckage of contexts that surrendered to or were overwhelmed by the new technology. And, while much preferable to unregulated new technologies, heavy regulation can bring its own ills.
If human beings are truly ready for AI and are not merely grasping at a promise of power and wisdom before the fitting time, the proof will be in our capacity to control it, to subordinate it to the proper ends of humanity, to prevent the practices of human beings and human societies from being overwhelmed by it, and jealousy to guard the goods of humanity from being sacrificed. As with the car, mere individual virtue is not going to be enough. In addition to extensive training and formation of individuals to use it responsibly (and to be permitted to use it at all), far-reaching regulation and restriction will almost certainly be necessary.
Questions of AI alignment with human beings are inextricable from broader political questions of human alignment with each other. And, even bracketing questions of AI alignment, so many dimensions of the crisis of AI result from and accelerate deeper problems of human “non-alignment,” from the distrust, alienation, fragmentation, and lack of consensus in human society. Like many other technologies, AI promises power against opponents and independence from others’ control and demands while correspondingly raising the risks we face. The technological bargains we have struck over the past couple of centuries have overwhelmingly favored and empowered individual autonomy against the common good and social demands. These bargains have not inexorably and inevitably followed from the technological innovations themselves but have been personal, societal, and political choices. AI presents us with further political decisions, among other things inviting us to press the logic of autonomy to a much greater extent. The key question is whether we are capable of making virtuous and wise decisions as societies and individuals, protecting human institutions and values, while availing ourselves of the affordances of new technologies circumspectly. Where we are not, we would probably be better off resisting the general adoption of this new technology until a future time when we might be.
Alastair Roberts (PhD, Durham) is adjunct Senior Fellow at Theopolis and is one of the participants in the Mere Fidelity podcast.. He blogs at Alastair’s Adversaria and tweets using @zugzwanged.
NOTES
[1] Abraham H. Maslow, The Psychology of Science: A Reconnaissance (Harper & Row, 1966) 15–16.