Is the AI “Revolution” the Latest Example of Technological Change Driving History?

Signed by Albert Einstein but written chiefly by the Hungarian physicist Leo Szilard, the letter is fairly considered the mother of all warnings about an emerging technology: in this case, the possible advent of atomic weapons, fueled by revelatory advances in Germany. Dated 2 August 1939, the letter begins calmly with Einstein reporting:

“the element uranium may be turned into a new and important source of energy in the immediate future. Certain aspects of the situation, which has arisen, seem to call for watchfulness and, if necessary, quick action on the part of the Administration.”

Einstein confessed next that he felt “it is my duty” to inform the President that, based on research in France and the U.S., “it appears almost certain [Math Processing Error] a nuclear chain reaction,” using “a large mass of uranium… could be achieved in the immediate future.” Einstein continued, “This new phenomenon would also lead to construction of bombs… extremely powerful bombs of a new type.” Einstein advised that Roosevelt forge “some permanent contact” with physicists and to support accelerated “experimental work” in cooperation with “industrial laboratories which have the necessary equipment.” In closing, Einstein noted that Germany, already at war in Europe, “has actually stopped the sale of uranium” from mines seized in what was then Czechoslovakia.

The Einstein letter can be viewed now as the mother of all warnings about an emerging destabilizing technology and suggests how warnings such as this one can set off a race to achieve the very engineered systems that humanity fears, on the assumption that the only defense against a revolutionary technology is to achieve effective dominance over it.

This ominous letter reached Roosevelt directly through astute aides, and then set off its own kind of chain reaction within the White House, which ultimately resulted in the formation of the Manhattan Project. Scientists and engineers gathered together in secret, nourished by massive resources provided by President Roosevelt and Congress, with the aim of outracing Germany in the design, testing, and production of an atomic weapon. In retrospect, the Germans never mounted a sustained effort, and after the U.S. detonated A-bombs on the defenseless Japanese cities of Hiroshima and Nagasaki, Einstein deeply regretted attaching his name to this formidable letter.

The Einstein letter can be viewed now as the mother of all warnings about an emerging destabilizing technology and suggests how warnings such as this one can set off a race to achieve the very engineered systems that humanity fears, on the assumption that the only defense against a revolutionary technology is to achieve effective dominance over it.

I introduce the Einstein letter in an attempt to illustrate that warnings about emerging technologies possess a history and adhere loosely to some patterns that, when stripped of details, amount to a warning paradigm, or model. The effective warning about a neglected or hidden technological threat has the following traits:

• The threat is presented in vivid, bold terms as immediate, urgent, and neglected at the peril of a nation or society.
• The threat is posed as transnational, with damage not limited to specific countries.
• The warning comes from a reliable or privileged source, a source that possesses special knowledge or awareness or both. Often, the warnings come from the very people who set out to build the technology they fear could harm them.

Warning against emerging perils from technological advance has become, in the years since 1945, has become normalized. So many warnings, about so many man-made perils, are issued that people risk becoming numb to them. This numbness gives rise to an alternative paradigm, described to me in an email by the philosopher of technology Carl Mitcham: the “ignoring paradigm.” While diverse in their scientific, engineering and bio-medical character, these warnings (whether activating or not) all share a basic perspective: they are deterministic; only forceful interventions, adaptations and management can limit or neuter the peril.

Fears of technological systems were a staple during the aftermath of World War II and into the 1960s. In Unsafe at Any Speed (1965), Ralph Nader eviscerated the safety of automobiles. In Silent Spring (1962), Rachel Carson documented the abusive use of chemicals in the environment. Barry Commoner, in Science and Survival (1966), drawing on his studies of the deleterious effects of above-ground testing of nuclear weapons, announced that “the age of innocent faith in science and technology may be over.” Michael Harrington, in The Other America (1962), presented a vivid picture of persistent poverty amid technological “progress.” In The Making of a Counter Culture (1969), Theodore Roszak connected youthful dissent with criticisms of “technocratic society.” By the 1970s, emerging technologies routinely ignited a sense of dread, helplessness, and anxiety that sometimes resulted in support for bans on certain technologies, most notably in the field of nuclear power. In a historic reversal, which many trace to the benign reception given “personal” technologies by the early advocates of micro-computers and other “tools” for better living (promoted by such popularizers as Stewart Brand’s Whole Earth Catalogue) emerging social technologies since the 1990s in have been cheered and enthusiastically embraced, at least at first.3

Perhaps the controlling example of the extreme enthusiasm for technological change was found in the advent of the World Wide Web in the mid-1990s. As personal computers heralded an expansion of freedom for the individual, so too did networks of PCs, which came to be known as the internet. While today the web is viewed as a mixed bag at best, and at worst the source of all forms of diminishment of human experience, at the outset the web was cheered as a promoter of fast, free retrieval of information, inexpensive mass communications, and boundless creative freedom. The political expression of the web’s benevolence was a slew of legislation during the presidency of Bill Clinton that effectively gave prime agents on the web a free pass from taxes, liability, and any forms of responsibility for the ill effects of content.4

The perception of the automatic benefits for the spread of the web, as with the earliest alarming descriptions of the potential destructiveness of atomic weapons, was cloaked in a presumption of inevitability. Oppose the web at your own risk. That the deterministic character of the web might include negative as well as positive aspects for people, places, and communities was neglected, if not utterly ignored. In his intimate account of how and why he invented the software “links” that animate the web as a system, Weaving the Web, Tim Berners-Lee identified his inspiration behind his seminal creation with an aha-moment he experienced in 1980: “Suppose all the information stored on computers everywhere was linked, I thought. Suppose I could program my computer to create a space in which anything could be linked to anything” [italics in original] [12, p. 4].

Captivated by his dream of “a single global information space,” Berners-Lee imagined the invention of “a web of information” in which humans would thrive at the center of a knowledge ecology of their own making:

“Computers might not find solutions to our problems, but they would be able to do the bulk of the legwork required, assisting our human minds, in intuitively finding ways through the maze. The added excitement was that computers also could follow and analyze tentative connective relationships that defined much of our society’s workings, unveiling entirely new ways to see our world. A system able to do that would be a fantastic thing–for managers, for social scientists and, ultimately, for everyone” [12, p. 5].

What could go wrong with such unfettered connectivity? Maybe the “genius” of human creativity is rooted in drudge work, in the tiresome toil of sorting through wheat from chaff with our own minds. Or perhaps less-educated, more provincial, and griev-ance-consumed individuals would harness these same tools for less fantastic ends. Berners-Lee would not open these doors of deception. He is not about to list the potential problems of surrendering parts of the terrain of human cognition to a machine. No worries about mental atrophy afflicting the high-minded or the ground-laying by our communal software routines promoting inaccuracies. A new age of expression will dawn, he predicts in 1999, when he publishes Weaving the Web. In the penultimate paragraph of his book, Berners-Lee exudes innocence and idealism about the future of an enhanced human community:

“Hope in life comes from the interconnections among all the people in the world. We believe that if we all work for what we think individually is good, then we as a whole will achieve more understanding, more harmony as we continue the journey. We don’t find the individual being subjugated by the whole. We don’t find the needs of the whole being subjugated by the increasing power of an individual. But we might see more understanding in the struggles between these extremes. We don’t expect the system to eventually become perfect. But we feel better and better about it. We find the journey more and more exciting, but we don’t expect it to end” [12, p. 209].

More harmony? More understanding? “Just better connected–connected into a better shape,” Bern-ers-Lee optimistically predicted. Apparently not. In the dream world of the web as imagined by Bern-ers-Lee, there are no fundamental disagreements that arise inexorably from different values and beliefs, aspirations, and material conditions; no insistent disputes over race, ethnicity, nationality, or gender. Perhaps among privileged global elites, Berners-Lee’s description is accurate, but in the billions of lives lived beyond the cloistered cosmopolitan world of the highly educated and well-to-do, the web appears to be captured by greed and grievance. And besides the dominance of false claims, fabrications, pornography, and financial scams, the capacity for the web to promote political repression and social control goes unmentioned. Furthermore, Berners-Lee fails to hint at the dark alternative to the harmonious “weaving” together of humankind. Alongside utopian visions of web-mediated self- and social-actualization, a “digital panopticon” is taking form, deterministically empowering governments and dictators to exploit the digital footprints of their citizens to constrain and monitor even intimate aspects of their lives.5

When a quarter of the way through the 21st century determinism remains so entrenched in public conversations about technological change, even by earnest policymakers and highly informed engineers and tech entrepreneurs; and when an air of inevitability surrounds expectations of the future effects of artificial intelligence and drone warfare, should we not pause and ask more fundamental questions about the nature of the intellectual pursuit of understanding the relation of science and society, of engineering and innovation, of technological systems and what humans want and need to thrive?6

How might we begin to engage these deeper questions? In my view, the starting place should be a re-engagement with the classic writings of Langdon Winner. Sadly, I have not spoken at length with Dr. Winner in some 40 years. I last saw him during my tenure at Arizona State University, when he came for a friendly visit to a faculty member in our science and technology studies unit. Because of his extended retreat from published debates on the logic and structure of technological change, I must call attention to the essay he published, some 25 years ago, in an unassuming but perspicacious book edited by professors Stephen Cutcliffe and Carl Mitcham, Visions of STS: Counterpoints in science, technology and social studies [15]. Dr. Winner’s essay, “Where Technological Determinism Went,” remains a vital point of departure for further critical thinking. Two paragraphs in this essay are especially salient. In the first, he writes:

“While new theories of social construction of technology emphasize multiple sources of innovation, numerous branching points, and continuing negotiation among social groups, a widely shared conviction about technological change – held just about everywhere else – is something quite different. It holds that people must scramble to catch up with developments whose course is, for all intents and purposes, beyond deliberation and judgement[Math Processing Error] .” [15].

Winner then adds, two paragraphs later, a rather harsh but not unsupported observation:

“I find it interesting that the scholarly community in STS is so inward looking that it seems not to notice the glaring disconnect between its own favored theories and visions of run-away technology that prevail in society at large. True, the new methods and models are useful for historical study–reconstructing choices that have already been made, speculating about how outcomes might have been different, and so forth. But as for ways to illuminate matters before us right now, the new notions (contingency, interpretive flexibility, actor networks, and the like) seem distinctly impotent” [15].

Twenty-five years have passed since Winner published these sobering words. Nothing fundamental has changed. Winner could speak these words now, this very day, and the same conditions—and contra-dictions—would hold sway. Is not it time to accept the challenge of a fundamental reconsideration?

As a starting premise, instead of thinking of determinism as an inevitably or an agent of imprisonment, perhaps we begin with a thought experiment: engage deterministic analysis in ways that will open doors of cognition now closed; and in doing so, maybe minds will open to the possibility that the constraints imposed by technological systems, whether hard or soft limits, merit greater attention.