by David Hoyt
Had the historian of California, Mike Davis, lived long enough to take in the spectacle of the AI boom, he very well may have seen a through line to earlier hybrids of unvarnished hype, greed, military research, science fiction, and techno-millenarianism, all perduring elements of a unique California ecosystem that has evolved some of the nation’s more distinctive cultural formations. If you think Peter Thiel’s itinerant, closed-door lectures on the Antichrist, Elon Musk’s high-volume procreation, or Sam Altman’s fascination with Napoleon and his notion of founding a company like you’re founding a religion were each a bit unusual, you have not read the history of California’s Church of Scientology.
Scientology – which has the same number of letters as “Singularity”- came into the world in the setting of a Pasadena club for Satan-worshipping libertines presided over by Caltech rocket scientist, sci-fi enthusiast and later founder of the Jet Propulsion Laboratory, John Parsons. As Davis tells the tale in his classic City of Quartz,
“Aside from being a world-famous rocket pioneer and a secret wizard, Parsons was also a devoted science fiction fan who attended meetings of the Los Angeles Fantasy and Science Fiction Society …One day in August 1945, to Parsons’s delight, a LAFSFS acquaintance showed up at [his] mansion with a young naval officer, Lt. Commander L. Ron Hubbard, who had already established a reputation as a master of sci-fi pulp.”
A romantic imbroglio then ensued between Hubbard and Parsons’s mistress, the outcome of which was a “vast diabolical experiment…to call up a true “whore of Babylon” so that she and Parsons might procreate a literal Antichrist in Pasadena.” Sometime later, failing to get rich writing for pulp sci-fi mags, Hubbard decided it would be more lucrative to invent a religion, “a cosmology derived from the pages of Astounding Science Fiction” and grounded in the “occult dramaturgy” picked up from Parsons.
One might say that, under the banner of the Antichrist, the so-called California Ideology that emerged with the privatization of the Internet in 1995 has managed to unify the long-divergent northern and southern cultures of the state into one cultural domain grounded in cyber-fabulism. A sign of this is the similarity between two websites from the late 1990’s and early 2000’s that continue to flicker in the ether like votive candles in a damp side chapel of a provincial Gothic cathedral. The “Heaven’s Gate” website (1997), a once-obscure UFO cult based in southern California, and the promotional webpage for Ray Kurzweil’s 2005 book, The Singularity is Near, promoting the sunny speculations of a Google machine learning engineer, were put up less than ten years apart. They share an utter conviction that things will automatically get better, but for different reasons and by different means.
For the Heaven’s Gate group, the moment of transcendence would occur when a “Mother Ship” arrived from outer space to extract believers (following their collective suicide) and elevate them to a higher, extraterrestrial level. For Kurzweil, writing in the San Francisco Bay Area, a selection of upward trends exemplified by Moore’s Law and seen as permanent, are extrapolated to the rest of human experience, such that the happy merger of humans and machines is claimed to be inevitable. This is the “singularity” taken up by CEO’s such as Microsoft’s Satya Nadella, following AI researcher Demis Hassabis, in the “foothills” of which we supposedly now stand.
This AI singularity is the Mother Ship of the artificial intelligence boosters. It is coming, the CEO’s tell us; it will be here soon as long as we can get more compute and venture capital. Then, as in the pages of Astounding Science Fiction, you will be amazed. Or, as OpenAI’s Sam Altman blogged with characteristic understatement in 2024, “This may turn out to be the most consequential fact about all of history so far. It is possible that we will have superintelligence in a few thousand days (!)” “[A]stounding triumphs – fixing the climate, establishing a space colony, and the discovery of all of physics – will eventually become commonplace.” Like Scientology before it (and like the animated television series The Jetsons, ca. 1962), the superintelligence singularity puts a smiley face on the future, ignores political conflict and diversity of all sorts, draws from a mash-up of science fiction at the level of a mediocre sophomore essay, and has the side benefit of helping a few folks make a lot of money.
In Ali Riza Taşkale’s recent essay, The Looting of Science Fiction, he looks at the uses of classic science fiction texts made by tech founders from Elon Musk to Jeff Bezos. The influence of sci-fi in its post-pulp era has been broad and acknowledged, but less so the liberties taken by tech founders with the more critical political and economic levels of the corpus. Across the board the tendency has been to keep the tech and mute the social critique of authoritarianism or inequality. What is presented in Isaac Asimov’s Foundation series, or Ridley Scott’s Bladerunner as speculative dystopia is repackaged as the objective present for which tech wielded by genius founders is the solution. As Taşkale writes,
Science fiction in the 1950’s imagined flying cars, abundant energy and more – but it did so before Three Mile Island, before Chernobyl, before we learned, often through disaster, what happens when you prioritize speed over safety…The optimistic aesthetic gets borrowed; the learning gets discarded.
From The Lord of the Rings to Palantir, or from Star Trek to the US Space Force, imaginative fiction that critiqued power or imagined an egalitarian, multi-racial society is appropriated in ways that endorse their opposites.
Both Heaven’s Gate and Kurzweil’s Singularity (the term is borrowed from cosmology and refers, comfortingly, to the edge of a black hole beyond which even light cannot escape) promise a transcendence of the human mediated by technology. Neither mention life beyond the human, the biological heritage of over three billion years of evolution, or the idea of ecological interdependence as anything other than hindrances or something to be left behind or improved. While Heaven’s Gate looked to salvation from alien races who would perform a rapture-by-abduction of elevated souls, the ideology of superintelligence and its moment of singularity pictures a material, earth-bound momentum towards transcendence based purely on the virgin-birth of technology. “There is a group of people,” an informant disclosed to researcher Karen Hao, “Ilya [Sutskever] being one of them – who believe that building AGI will bring about a rapture. Literally, a rapture.”
The passive posture of Heaven’s Gate, with its appeal to isolated individuals of salvation from somewhere else, resonated with the besieged vibe of the 1970’s, a period which many at the time felt was traversing a moment of polycrisis not unlike our own, in which the social contract seemed to be fraying and advanced liberal democracies had become ungovernable. Heaven’s Gate spurned money and materialism. Yet they were early adopters of the Web and felt that advanced technology was a mark of superior morality. As stated on their website, “The Next Level – the true Kingdom of God – has the only truly advanced space-time travel vehicles, or spacecrafts.”
What makes the cult of AI different in tone and in worldview is its origin within a tech industry which sits at an epicenter of unregulated flows of global capital, its profound, longstanding, and growing links to the US security state, and the certainty that on the basis of these pillars it can build the future it wants here on earth right now, and that no one should be able to stop them.
On the terrestrial level of reality, AI’s colonial dynamic is embodied in the surreptitious construction of enormously expensive, climate-damaging and resource-intensive data centers world-wide, often with disregard for strong local opposition. It is embodied in xAI’s Colossus data center outside Memphis, Tennessee, and another near Southaven, Mississippi, which both run on unpermitted and polluting gas turbines situated near poor Black communities. It is embodied in Musk’s vision of a cyborg and robot-filled future, which is now contractually tied to his corporate compensation, and anchored in what is planned to be the largest semiconductor factory in the world, Terafab, also powered 100% on natural gas. If Sam Altman expects AI to “solve climate change” some day, he had better get to it, because everything that is being done to accelerate AI is turning up the heat on planet Earth.
Two macroeconomic factors are at play in the breakneck speed with which AI infrastructure has been financed and built out since the release of Chat GPT in November of 2022. They have little to do with the intrinsic commercial potential of AI-related products or services, nor any particular innovation in computer engineering. The first, to put it simply, is the availability of lots of money. More than ever before, giant pots of money are sloshing around the planet, motivated by historically low interest rates over a relatively long period to search out the Next Big Thing. This money is mostly exempt from national regimes of control or taxation, is managed outside the traditional banking system, and is therefore more free to move around. The second is the chronic and by now widely acknowledged slow-down of economic growth in the advanced economies, dating back nearly half a century. What the hype typically fails to register is that investors don’t seek out technology and AI firms because they are the best bets on future returns – they seek them out because they are the only bets around.
To understand just why this is, Nick Srnicek provides a useful capsule history of the sequence of financial crises that have followed the liberalization of capital since the later 20th century. It is a story that leads with a crescendo to the current boom in AI development. In the mid 1990’s, the hottest global market was not centered in Silicon Valley, but in the small economies of East and Southeast Asia, the “Asian Tigers.” Only when these blew up beginning in 1997 did newly mobile capital, recently freed from decades of local restrictions on its movement in and out of national economies, begin to pour into Silicon Valley in what would become known as the dot-com bubble. When this blew up, as all bubbles do, it led to a decade of low interest rates dictated by the Federal Reserve. This, in turn, led global investors to flood the US housing market, inflating a nearly decade-long bubble in real estate prices. When this bubble blew up in 2008 with even more severity, followed almost immediately by the Greek/European debt crisis of 2009-2012, the solvency of major financial institutions in the US and Europe and of the entire economic order was threatened. Only massive intervention from the public sector avoided a meltdown. At the same time, the year 2012 marked the year that growth rates in the People’s Republic of China dropped under 10% annually for the first time in over a decade. This twin shock to the global economy, in combination with austerity measures put in place in the aftermath of the 2008 crisis, led to the unraveling of the global economic architecture put in place over the previous quarter century.
This is the context of chronic low growth in which money has been pouring into AI, either invested in privately held firms such as Anthropic or OpenAI, or publically held tech firms listed in the US such as Google and Microsoft, or in overseas firms such as semiconductor giant SK Hynix in South Korea. The search for returns continues to drive valuations of AI majors to astronomical heights, bringing significant portions of the market along with them. In May, 2026 Anthropic, which only recently turned a modest profit on sales of its AI assistant chatbot Claude, came close to breaking the valuation threshold of $1 trillion USD. (There is no standard, objective methodology for obtaining such valuations for private companies). According to the OECD, artificial intelligence firms captured 61% of global venture capital in 2025. A Stanford report puts global VC investment in Silicon Valley firms alone at 75% of the total. Overall capital spending on AI in the United States is at roughly 2% of GDP, and expected to reach 3-4% in 2027. This is equal to and slightly exceeding the US defense budget.
This is a lot of investment, but in what, precisely? As it stands right now, AI is an all-purpose everything enhancer, like the patent cure-all medicines of the nineteenth century. No one knows exactly what it is for. Given that it’s not at all clear what “intelligence” is, let alone human intelligence, it’s not surprising that “artificial intelligence” is a rather catch-all category (of a distinctly old-school, Cartesian sort) that can mean everything from the technology used in self-driving cars or a chatbot whispering sweet nothings or Nazi propaganda. It can mean something that can sort through options within a closed system, like how to lay out an architectural floor plan or where to place turbines on a wind farm. It’s also really good at playing ancient Chinese board games. But none of these capabilities are invested in directly. What’s invested in is compute, which means data centers, which means steel and cement and copper and chips and routers, and all of the various ancillary industries that support AI infrastructure.
To this end, on Monday, August 10, 2026, the business press announced that Wall Street, through a consortium of global asset management titans such as BlackRock and chip maker NVIDIA, was setting up a $500 billion fund for low-interest loans to AI startups which were having a hard time financing the very expensive NVIDIA chips requisite for any AI undertaking. This is on top of the seemingly endless and ever-larger sums pledged by the big tech firms over the course of fiscal 2026 ($145 billion on data centers from Meta; $205 billion for data centers from Alphabet/Google; $321.9 billion from Microsoft for data center leases in fiscal 2026, with $13 billion directly to OpenAI).
What marks a turning point in the August, 2026 BlackRock announcement is the shift in AI financing, from the vast cash reserves long held by each of these companies, to debt financing. Amid early warnings that the AI data center build-out might be a speculative bubble reminiscent of the dotcom frenzy of the late 1990’s, analysts noted that much of the capital expenditure was drawing down the enormous cash piles of established, profitable companies, and that comparison to the startup ecosystem of the dotcom bust was therefore unwarranted. Any risk involved in the build-out was would be insulated from the larger economy. In strictly financial terms, this was true.
Until now.
In Alphabet’s second quarter 2026 earnings call, and despite Alphabet’s record cloud revenue, its projected 2026 capital expenditures and newly negative net free cash flow sent its stock tumbling, erasing billions of dollars of capitalization in a few hours. The markets are clearly nervous that all of this spending might not pay off, alternating between euphoric optimism and repeatedly issuing caveats that a lot of Claude or GPT chatbot agents will need to be sold to pay for all of this, that the build-out has now exceeded the cash reserves of Big Tech, and that the growing costs will henceforth be financed by the likes of BlackRock, which stands a good chance of holding a chunk of the money in your retirement or pension account.
If all of this sounds strangely familiar, it should. It was not that long ago that, in the immediate aftermath of COVD lockdowns and growing concern with global warming, the Biden administration passed historic legislation to advance the so-called “green transition” and decarbonization of the economy. Then, too, and understandably, the impulse was to move fast. But in that instance the goal was clear, in contrast to the present moment: to hold down average global surface temperatures to within limits that would not destabilize the climate. It’s useful to read what Larry Fink, BlackRock CEO and elder statesman of Wall Street, had to say in his widely read Letter of 2022, after passage of the Inflation Reduction Act and the Infrastructure Investment and Jobs Act.
“Engineers and scientists are working around the clock on how to decarbonize cement, steel, and plastics; shipping, trucking, and aviation; agriculture, energy and construction. I believe the decarbonizing of the global economy is going to create the greatest investment opportunity of our lifetime… The next 1,000 unicorns won’t be search engines or social media companies, they’ll be sustainable, scalable innovators – startups that help the world decarbonize and make the energy transition affordable for all consumers.”
Fast forward nearly five years, and Oh, how things have changed! BlackRock is no longer nudging corporate opinion on the perils of global warming, though the perils of global warming are making themselves apparent without help from Larry Fink. (July 2026, as anyone living in Europe, or northern Canada knows, was the hottest in recorded history). This underlines how easily and quickly investment sentiment can shift. It also illuminates the political and macroeconomic contours within which the AI boom, and the technology behind the preferred model of artificial intelligence (based on hyperscaling of large language models), have taken shape.
The nineteenth century British inventor and mathematician Charles Babbage, considered to be the father of the digital computer, saw machines, including his own “difference engine”, as mechanical compressions of existing social divisions of labor. In a similar way, AI is the digital compression of unregulated platform capitalism, the very business model of which is to get big, achieve network effects, and collect rent. It depends on human data amassed with little consent, which is used to modify behavior in a way favorable to the growth of the companies deploying it.
If Larry Fink no longer sees global warming as an investment opportunity, if Sam Altman thinks AI “may help” solve climate change some day, this is because we now live in a moment where collective decisions affecting the global community and ecological health are not being made at the level of representative governments or executed by public power, but by asset management firms and billionaire CEO’s in thrall to the latest growth play or to their particular and occasionally reactionary futurologies. It is no revelation to state that AI is not broadly popular with the American public, that data centers are widely despised, and that there is no love lost for the restricted class of wealthy individuals who are rapidly and without consultation privately terraforming the commons in a moment of ecological crisis. This is because the public, in a rare moment of wisdom, clearly understands what is going on.
There is also the possibility that AI is not the Next Big Thing, the long-sought engine of economic growth equivalent to that of the Gilded Age with its railroads and robber barons. The quantum leap in economic productivity that has eluded forecasters for the last half century may, as Robert Gordon and Vaclav Smil argue, have been historically unrepeatable. When this bubble bursts, as they always do, we may thus all be left to pay for a whole lot of infrastructure, and a greatly enhanced surveillance state.
A future Thucydides, perhaps a lonely chatbot answering self-generated queries, may well look back on this time, when the living, biological heritage of eons was in convulsions before our eyes, and wonder how it was that instead of touching grass, we spent our gold on unearthly science fictions.