Before a single steel beam gets welded, a factory needs one thing.

Power.

Before an assembly line turns, a refinery runs, or a new plant opens its doors, someone has to generate the electricity behind it.

And in Texas, the demand for that electricity is climbing fast.

ERCOT projects roughly 70% electricity demand growth by 2031.

More than 430 GW of new power demand has been requested in the state. That is nearly five times Texas's current peak demand.

That gap won't close by itself. Someone has to build the power plants.

American PowerGen is one of the companies developing them.

It is developing an estimated 22 GW pipeline of natural gas power generation across 16 Texas projects, in various stages of development.

For its two most advanced projects, much of the hardest part is done, as described in the offering materials:

More than 1,000 acres under option.

Natural gas availability identified.

Grid interconnection agreements negotiated.

Air permits secured.

Behind it is a leadership team with 12 power plant exits.

American PowerGen has also reserved the Nasdaq ticker "APGX," which does not indicate that a listing is likely.

Today, shares are $2.50, with a $1,000 minimum.

Early investors can receive up to 15% bonus shares.

The round closes at 11:59 PM ET on Wednesday, November 18, 2026.

Lock In $2.50 Per Share Today

Disclaimer: This is a paid advertisement for American PowerGen's Regulation Crowdfunding offering. Please read the offering materials and Form C at https://invest.americanpowergen.com/

Reservation of a ticker symbol does not indicate that the Company's securities are currently listed or traded on Nasdaq, does not constitute approval by Nasdaq, and is not a guarantee that the Company will be listed on Nasdaq. Any future listing remains subject to Nasdaq's applicable listing standards, regulatory requirements, and approvals.

Equity crowdfunding investments in private placements, and start-up investments in particular, are speculative and involve a high degree of risk, and those investors who cannot afford to lose their entire investment should not invest in start-ups.

Shares purchased in this offering are subject to a one-year lock-up before they may be resold, subject to limited exceptions. Shares are not publicly traded. The company's stated expected exit timeline is 5-10 years, via acquisition or IPO, and no exit is guaranteed.

Early Investor Bonus shares are offered for a limited time, are cumulative, and are capped at 20%. Bonus share tiers are subject to the terms of the offering.

This communication may contain forward-looking statements relating to the company, its business plan, and its strategy. Forward-looking statements are subject to material risks and uncertainties, and actual results may differ materially.

DealMaker Securities LLC, a registered broker-dealer and member of FINRA | SIPC, located at 30 East 23rd Street, 2nd Floor, New York, NY 10010, is the Intermediary for this offering.

A metal most people have never heard of nearly tripled in price in under a year. Hafnium averaged $14,200 per kilogram in the second quarter of 2026. A year earlier it sat near $5,000. No new mine can fix this. No surge in production is coming. The price moved because the supply is physically trapped.

Hafnium cannot be mined on its own. It exists inside zirconium ore at a ratio of about 50 to 1. To get one tonne of hafnium, refiners must process 50 tonnes of zirconium. Total global output sits at 70 to 75 tonnes a year. Four countries produce it: France, the United States, China, and Russia. That is the entire world supply.

Hafnium's price is not moving on speculation. Three demand forces are pulling from a supply pipe that cannot widen. All three are tied to the AI era. AI chips need hafnium. The gas turbines powering AI data centers need hafnium. Nuclear reactors need hafnium. All three streams draw from the same 70-tonne annual supply. None of them are shrinking.

An email announcing a new 401(k) investment option is coming to 70 million inboxes. Most will delete it – and miss what Larry Benedict calls the decade's biggest opportunity. He's naming the one ticker to own before it arrives – free. Watch here.

Hafnium has no dedicated mine anywhere on Earth. It rides along with zirconium. Refiners separate the two during processing, but only when there is demand for nuclear-grade zirconium. The U.S. Geological Survey classifies hafnium supply as set by zirconium production, not by hafnium demand.

This means one thing. The price can climb forever. No one can dig more hafnium out of the ground. The constraint is physical. Zirconium demand sets the ceiling. If zirconium production stays flat, hafnium production stays flat. Price signals that would normally trigger new mining do nothing here. The valve is welded at one width.

Observation: Global hafnium supply is fixed at 70 to 75 tonnes per year, with no independent production path.
Interpretation: Supply cannot respond to demand. The market has no release valve.

The first stream is semiconductors. Hafnium dioxide is the insulating layer in the gate of every advanced transistor. Intel introduced it at the 45-nanometer chip node in 2007. Every major chipmaker followed. As AI chips pack more transistors per wafer, each wafer eats more hafnium. Semiconductors already account for roughly half of all hafnium demand in 2025. The Minor Metals Trade Association puts semiconductor consumption at 40 tonnes a year today. By 2030, that figure reaches 64 tonnes. That single stream would consume nearly the entire global supply.

The second stream is power. AI data centers need electricity. They are buying gas turbines faster than anyone can build them. Hafnium goes into the superalloy blades that survive extreme heat inside those turbines. Siemens Energy sold 194 gas turbines in 2025. About 65 percent went to data centers. The demand runs beyond Siemens. GE Vernova ended Q2 2026 with 116 gigawatts of turbine orders. Reuters reported in November 2026 that the global backlog exceeds 100 gigawatts. Manufacturers can produce only 60 to 70 gigawatts a year. Every turbine waiting in that gap needs hafnium. The AI boom demands hafnium twice: once for the chip, once for the power to run it.

The third stream is nuclear energy. Hafnium absorbs neutrons. That property makes it essential in reactor control rods. Japan and South Korea have both increased purchases to support higher reactor output. The volume is smaller than chips or turbines, but it adds to the same fixed pipe.

A trader told Argus there are 12 tonnes of excess demand the market cannot meet. Against a 70-tonne supply, that is a 17 percent structural shortage. All three demand streams are still growing.

Observation: Semiconductor, turbine, and nuclear demand are converging on a fixed 70-tonne supply, with 12 tonnes of excess demand already unmet.
Interpretation: The shortage is structural, not cyclical. Each demand stream is tied to infrastructure buildouts that will take years to complete.

China tightened export controls on hafnium in 2024. The regulations cover dual-use materials, items with both civilian and military applications. There was no formal ban. The government imposed stricter licensing on exporters. The effect was immediate. Unwrought hafnium exports stood at 5,001 kilograms in January 2025. By September 2025, they had fallen to 499 kilograms. That is a 90 percent drop in nine months.

A metal trader told Argus how the system works. "Very few parties can get their licenses granted." Without a track record of handling controlled materials, companies simply do not receive them.

China's own demand is growing. Its nuclear expansion gives it reason to keep hafnium at home rather than ship it abroad. The licensing system gives Beijing a quiet lever. It can restrict supply without a formal announcement.

The U.S. lists hafnium under Title III of the Defense Production Act. Title III authorizes federal funding to expand domestic production of critical materials. But as of mid-2026, no hafnium-specific award has been disclosed. The material is on the list. The money is not.

Observation: China's hafnium exports fell 90 percent in nine months through licensing restrictions, not a formal ban.
Interpretation: The supply side is not just structurally trapped. It is being actively tightened by a major refiner, with no U.S. production response funded.

Three signals are worth tracking in the weeks ahead. First, semiconductor hafnium consumption. The trajectory from 40 tonnes toward 64 tonnes by 2030 is the largest demand variable. Any acceleration in advanced chip production tightens the pipe further.

Second, turbine order backlogs. The gap between orders and production capacity shows the pressure on superalloy supply. That gap is currently 30 to 40 gigawatts and widening.

Third, Chinese export licensing data. The 90 percent decline happened without a formal ban. A full ban would remove the remaining trickle. The licensing mechanism lets China adjust supply without a public announcement. That makes the monthly customs figures the only reliable signal.

This is not a story about one commodity spiking. It is about what happens when three demand forces hit a supply pipe that cannot bend.

Hafnium's supply is locked. It is a byproduct at 50 to 1, with no mine and no funded path to expand output. Three AI demand streams pull from the same pipe, and China is tightening the valve.

Until next time,
The Navigator