When I say artificial intelligence doesn’t kill people, LLMs do, the objection I hear most often is that this is a semantic quarrel: that “AI” and “large language model” are near enough as to be interchangeable, and insisting on the difference is pedantry.
asserted
insisting → say → difference
Consider Boxtown, a neighborhood in Memphis, Tennessee, founded after emancipation by formerly enslaved people who built their first houses from discarded railroad boxcars.
asserted
who → consider → boxcars
Residents live alongside an oil refinery, a steel mill, and chemical plants, and report asthma and cancer rates far above national averages, all of it predating what came next.
asserted
what → live → it
What came to Boxtown next was a supercomputer.
asserted
came → come → Boxtown
In 2024, xAI built Colossus there, over 230,000 graphics processing units, in 122 days, and powered it with dozens of methane gas turbines for which it did not have permits.
asserted
it → build → permits
It was only caught breaking the law when the Southern Environmental Law Center flew thermal-imaging drones over the site and counted 35 of them, roughly enough capacity to power 280,000 homes.
asserted
Center → catch → homes
The Shelby County Health Department later issued Clean Air Act permits covering 15 of them, a decision environmental groups have appealed.
asserted
groups → issue → decision
In April, the NAACP sued xAI and a subsidiary over a second facility across the state line in Southaven, Mississippi, alleging 27 more unpermitted turbines emitting more than 1,700 tons of nitrogen oxides, 180 tons of fine particulate matter, and 19 tons of formaldehyde annually, which the complaint contends would make it the largest industrial source of nitrogen oxides in the greater Memphis area.
asserted
it → sue → area
The allegations have not been adjudicated.
asserted
allegations → adjudicate → ?
Nothing about artificial intelligence required that.
asserted
Nothing → require → that
The difference between “AI” and “LLMs” is measured in gigawatts, aquifers, and the air over a neighborhood that has been absorbing other people’s industry for a century.
asserted
that → measure → century
The evidence that settles it does not come from nostalgia about simpler computing.
asserted
that → settle → computing
It comes from what the best AI laboratories in the world have published in the last few years.
asserted
laboratories → come → years
Start there, because it reframes everything that follows.
asserted
that → start → everything
This year a team including Yann LeCun, a Turing Award laureate and one of the architects of modern deep learning, published a world model trained end to end on a single GPU in a few hours.
asserted
team → include → hours
It plans roughly 48 times faster than systems built on foundation models, using around 200 times fewer tokens.
asserted
It → plan → tokens
LeCun’s research program is, explicitly, an argument that scaling language models is the wrong road.
asserted
scaling → scale → models
And the man himself left Meta for good in November 2025, calling LLMs a “dead end.”
asserted
man → leave → LLMs
DeepMind’s GraphCast, published in Science, generates a 10-day global weather forecast on one tensor processing unit in under a minute and beats the European gold-standard supercomputer forecast in more than 90% of tested variables.
asserted
GraphCast → publish → variables
FourCastNet, from the group that pioneered neural operators for physics, is estimated to use 12,000 times less energy than the numerical weather model it substitutes for.
asserted
it → pioneer → model
That is artificial intelligence that reduces the world’s computational load.
asserted
that → reduce → load
A neural network built by DeepMind and the Swiss Plasma Center controls all 19 magnetic coils of a tokamak fusion reactor, in real time, running on the reactor’s own control hardware.
asserted
network → build → hardware
It was published in Nature.
asserted
It → publish → Nature
Confining fusion plasma is a harder real-time problem than holding a conversation, and it does not require a campus in Loudoun County, Virginia.
asserted
it → confine → County
At the other end of the scale, TinyML models run on microcontrollers drawing under a milliwatt, in less than 512 kilobytes of memory.
asserted
models → run → memory
ABI Research projects 2.5 billion devices shipping with that capability by 2030: billions of AI deployments that will never contact a data center at all, running for years on a coin cell.
asserted
that → project → cell
A commercial counterfactual is also emerging that receives almost no attention in this debate: firms building models the opposite way: a purpose-built architecture for a single task, trained on a customer’s own data, deployed on the customer’s own hardware, on-premise or air-gapped.
asserted
firms → emerge → premise
The model runs where the data already are.
asserted
data → run → ?
Every enterprise task served this way is an inference that never enters a hyperscale facility.
asserted
that → serve → facility
Now set that against what is actually being constructed.
U.S. data centers consumed roughly 4.4% of the nation’s electricity in 2023 and are projected to reach 6.7%-12% by 2028, according to Lawrence Berkeley National Laboratory, which attributes the doubling of data center demand between 2017 and 2023 largely to AI servers.
uncertain
which → set → servers
Put another way, one-ninth of all electricity consumed in the United States will go to powering these facilities by the end of the decade.
asserted
ninth → put → decade
The International Energy Agency has even more damning figures:
asserted
Agency → have → figures
Electricity use in accelerated servers, the GPU-dense racks that train and serve large models, is growing about 30% a year against 9% for conventional servers, and accounts for almost half the net increase in global data center consumption.
asserted
that → train → consumption
The compute used to train frontier models has grown four to five times per year for over a decade.
asserted
compute → use → decade
But it’s not the data center hyperscalers picking up the electricity tab, by and large.
asserted
it → ’ → tab
Rather, the bill arrives at U.S. households.
asserted
bill → arrive → households
In the PJM Interconnection, serving 65 million people, capacity prices rose from $28.92 per megawatt-day to $329.17 in two years.
asserted
prices → serve → years
Data centers were responsible for 63% of one auction’s increase, about $9.3 billion recovered from ratepayers, with the Natural Resources Defense Council projecting roughly $70 a month in additional household costs by 2028.
asserted
Council → recover → 2028
In that same record auction, PJM fell 6,625 megawatts short of its own reliability target for the first time in the capacity market’s history.
asserted
PJM → fall → history
A Bloomberg analysis of some 8,000 facilities found that about two-thirds of new U.S. data centers built or in development since 2022 sit in areas of high water stress.
asserted
thirds → find → stress
…and 31 more, not listed.
The Aporia
