Prompts From the End of Time

It was a good run

The End of The World, John Martin, c.1851-53

As the world hurtles toward climatologic and ecologic catastrophe, with many of its nations’ leaders concluding that hitting the accelerator on capitalism’s infinite growth mandate is the best course of action, let’s take a step back and examine the single greatest technological menace to life on Earth.

We’ve covered enshittification before, and it shows no signs of abating or reversing course. Generative AI, however, is not merely a byproduct of enshittification: it actively foments and enshittifies by its very nature.

Artificial intelligence has become the primary growth engine of the American economy. Harvard economist Jason Furman claims 92% of GDP growth from the first half of 2025 is attributable to AI investment.

While the idea that a massively overhyped, still unproven technology underpins the majority of our economic growth is disturbing, the series of events harkens back to the dotcom bubble of the late ‘90s.

The parallels are eerie: massive, cross-industry investments; ceaseless marketing and integration of AI features into almost every tech product we use; eye-watering valuations of AI companies that haven’t turned a profit yet. There is no shortage of analysts who don’t believe AI companies even have a path to profitability.

The desperately absurd energy requirements and carbon footprints of new AI datacenters likewise puzzle and horrify: OpenAI recently struck a deal with Broadcom to develop 10 gigawatts of additional AI datacenter capacity, roughly the energy usage of a large American city.

Naturally, AI’s purveyors see nothing but upside here. Additional capacity will enable amazing, incredible features that make your life as a faceless consumer of AI slop better. And, most critically, additional investment will make early investors in AI companies like OpenAI and Anthropic even richer. That’s the headline here, in case you missed it.

Sam Altman himself, CEO of OpenAI, is never very long on detail when it comes to just how AI will change the world, or how it will do so in a sustainable or ethical way, but he’s sure it will. Never mind that his company’s flagship product only exists because of large-scale theft of creative works produced by actual human beings.

And don’t pay any attention to the fact that the world’s largest insurance companies can’t insure the likes of OpenAI because the scale of potential (and pending) legal claims against those companies is so grand. 

Meanwhile, Elon Musk’s ever-hallucinatory, increasingly racist and Fascist AI model, Grok, now runs on new xAI datacenter infrastructure in Memphis, Tennessee where its nitrogen oxide (NOx) emissions are estimated at 1200 to 2000 tons per year—vastly higher than neighboring fossil-fuel-powered interests. The xAI datacenter is located in an area of South Memphis already known for its high smog levels and state-leading asthma hospitalizations.

Politico reports:

The turbines are only temporary and don’t require federal permits for their emissions of NOx and other hazardous air pollutants like formaldehyde, xAI’s environmental consultant, Shannon Lynn, said during a webinar hosted by the Memphis Chamber of Commerce. The argument appears to rely on a loophole in federal regulations that environmental groups and former EPA officials say shouldn’t apply to the situation.

Given Elon Musk’s hot-and-cold relationship with current President and convicted felon Donald J. Trump, it may not be too surprising that the xAI datacenter in Memphis retains carte blanche to pollute with impunity, ignoring federal regulations and the Clean Air Act in the process.

EPA Administrator Lee Zeldin has championed the deregulation of environmental protection standards, claiming the agency’s goal is now:

“driving a dagger straight into the heart of the climate change religion to drive down cost of living for American families, unleash American energy, bring auto jobs back to the U.S. and more,”

This hand-in-glove arrangement between the current regime in Washington and the AI-focused companies who have so generously supported them comes at a time when political capital should be spent on driving down emissions, not increasing them. Of course, Sam Altman and Elon Musk’s goals, and the goals of the AI ouroboros, are incompatible with such ‘strict’ emissions standards; if tech continues to have its way, we will likely only see emissions targets further reduced.

I appreciate that this topic is fraught and politically charged; AI is being deployed at a scale that precious few had any inkling of even five years ago. But this is rather the point: the longer we refuse to confront the deep interrelations between AI, policy, and our own environment, the closer we walk to cataclysm. Granted, this isn't the bubbly, "I support technological advancement at literally any cost" pseudo ad copy you'll read elsewhere on the Internet, but I don't have affiliate links and I don't make money from lying to you.

Generative AI’s growth is unsustainable; its promises are largely unfulfilled; our economy’s dependence on its success leaves us open to existential threats to our current way of life; the politics encouraging the AI bubble are diametrically opposed to the politics needed for a habitable planet; the privacy and safety implications of AI models without actual guardrails are dystopian; AI as a tool can and will be used by hostile governments to identify and punish dissenters.

For all its potential, AI is a loaded gun and those most prominently and publicly championing its rapid adoption are confidence-men at best, and deeply malicious, narcissistic grifters at worst.