Seems AI Needs to Use Unconventionally Generated Electricity If the Industry Wants to Thrive

In the Bible, we are advised to count the costs before diving into faith. In my 25+ years of being an executive sponsor, it was one of the first things my team did for any project. So what happened?

May 28, 2026

via Fortune, 22 May 2026 / Laura Aboli, TG

The company that bet its future on AI just told 100,000 engineers to stop using its best tool because it was bleeding them dry

Something very interesting is beginning to emerge from inside the AI industry itself.

For the past two years we have been told that AI would replace human workers, dramatically reduce costs and create unprecedented efficiency across every sector of the economy.

Markets soared on that promise.

Companies fired staff, announced “AI integration” and watched their stock prices rise accordingly.

But now some of the first major cracks are beginning to appear in the narrative.

Microsoft has reportedly started canceling large numbers of internal Claude Code licenses after costs spiraled far beyond expectations as engineers increasingly relied on the system.

Uber executives admitted their AI budgets were effectively blown apart within months of deployment.

Even Nvidia’s own VP of Applied Deep Learning openly stated that for his team, the cost of compute had become “far beyond the costs of the employees.”

What is becoming apparent is that large scale AI deployment may not actually reduce costs in the way investors were led to believe.

Quite the opposite.

The more powerful these systems become, the more they are used.

The more they are used, the more tokens, processing power, energy, infrastructure and compute they consume.

And at enterprise scale those costs become enormous.

The assumption was that companies would replace expensive humans with cheap AI.

Instead, they may end up needing: expensive humans supervising extremely expensive AI systems running on staggeringly expensive infrastructure.

AND THAT CHANGES THE ECONOMIC EQUATION ENTIRELY…

from Sir Escanor on X

“CEOs are quietly realizing the AI replacement plan has a problem.
Two problems, actually.

”One: the token costs for running AI agents are now exceeding what they were paying the employees they fired.

”Two: when the tokens run out, the AI stops. Just stops. No continuity. No workaround. Just a spinning wheel where your workforce used to be.

”You fired humans to save money and bought a subscription that bills you into a corner.

”The employees you let go knew what to do when things broke.

”The AI just invoices you for the outage.

”And then there’s the permission problem nobody wants to talk about.

”To do its job, the AI agent needs access.

“Full access.

“Your systems, your patents, your contracts, your future plans.

“Everything you spent years building, handed over to a process that has no loyalty, no discretion, and no skin in the game.

”You didn’t hire a replacement.

“You gave a stranger with no soul the keys to everything you own.

”Enjoy.”

from Professor Zenkus on X

“When CEOs replace employees with AI, they are signing up for a subscription service that - after whatever introductory price agreed to has expired - has the power to raise the cost of that service at will.

“And if the company can't pay, everything STOPS.

”Corporations are "hiring" AI bots who can in effect set their own salaries.

“And don't think for a minute that Big Tech won't raise the price as soon as they know you are dependent on them.

“But don't worry, because if the company cannot afford the ever rising costs for their new AI "workforce", I'm sure Anthropic or whatever AI firm they are contracted to will be happy to take part ownership of the company in exchange for continuing the service.

”Corporate America is writing its own death warrant.”

Our Note: the last statement is totally reliant upon the environment in which one is operating - there are ALWAYS options underfoot, one just has to simply raise their foot and search for those options, even if only one is apparent…

So Now What?

“… think in terms of energy, frequency, and vibration.” - Nikola Tesla

The master of free energy…

Many are still thinking of energy that has to have a meter.

ALSO…

Scalability and a wide, very wide, spectrum of shiftability are EXTREMELY important right now - otherwise today’s leader is tomorrow’s loser, quite literally in position and timeline.

Few are even looking at nuclear energy - FUSION not fission…

Nuclear fusion technology is experiencing rapid advancements, with significant progress made in recent years, drawing substantial private investment.

More than $7.1 billion has been raised for nuclear fusion research, and the private sector is increasingly involved in developing fusion reactors.

The goal is to achieve a working fusion reactor within the next decade, with the SPARC reactor expected to be completed in 2025.

These developments are leading to more compact reactors suitable for distributed utility grids, with some experiments demonstrating proof of concept and even achieving net positive energy output, meaning they produce more energy than required to run the reaction.

Such breakthroughs are pushing the field closer to commercial viability, although scalability and cost remain challenges.

Both public and private entities are collaborating to accelerate progress, with entities like Commonwealth Fusion Systems, General Fusion, and TAE Technologies making strides in this area.

Recent scientific and technological advancements include Type 1 Energy's stellarator design, which uses powerful magnetic fields to contain plasma heated to approximately 100 million degrees Celsius and plans to construct a prototype, Infinity One, by 2026.

This design aims to use abundant fuels like deuterium and lithium, producing harmless helium as waste - meaning there is no radioactive waste…

Breakthroughs at facilities like the National Ignition Facility have achieved ignition, where fusion energy produced equals the energy input from lasers.

Furthermore, Chinese researchers using the Experimental Advanced Superconducting Tokamak (EAST), also known as China's "artificial sun," have overcome a long-standing plasma density barrier, keeping plasma stable at extreme densities.

Artificial intelligence (AI) and quantum computing are playing crucial roles by assisting in the control and design of fusion reactors, adjusting magnetic fields, filling in missing data, and identifying safe zones within reactors.

The U.S. government is also investing in supercomputers to accelerate progress in fusion energy.

Numerous initiatives are underway globally, including the establishment of the Laboratory for Materials in Nuclear Technologies (LMNT) at MIT in June 2025, dedicated to finding cost-effective material solutions for commercial fusion reactors.

The UK Infinity Fusion Consortium, formed by Type One Energy, Tokamak Energy, and AECOM, is pursuing the development of the first private-sector-led fusion power plant in the UK.

The European Union has also allocated over €330 million for 2026–2027 to advance fusion research.

The U.S. Department of Energy (DOE) has developed a comprehensive Fusion Science & Technology (S&T) Roadmap to accelerate commercial fusion energy by the mid-2030s, fostering public-private partnerships.

While the timeline for widespread commercial deployment remains a subject of optimistic versus realistic projections, the overall outlook indicates substantial progress and a growing global commitment to harnessing fusion power.

sources:

Decisive Liberty

Discussion about this post

Ready for more?