In 1945, the U.S. Army tried something new. Instead of asking what a firm planned to build, they asked what it had already shipped. Not drawings, not test runs — finished goods, checked and signed for. The Army called it "proof of shipping." It set the bar for every defense contract since.

Jack Lifton wants rare earth firms to meet that same bar. In his September 16 column for InvestorNews, he went firm by firm through every Western project that claims it can produce dysprosium and terbium. Those are the two heavy rare earths that stop magnets from failing in high heat. His finding: nearly everyone has them in the ground. Nearly nobody can ship them.

One firm passes the test. Lynas produced its first separated dysprosium from its Malaysia plant in May 2025 and terbium in June. They have kept selling. That is the line. Every other project sits between "pilot sample" and "we plan to start next year." A two-kilogram lab batch and a working refinery are both called "production," but they are not the same feat.

The next group says 2026. MP Materials aimed for mid-year on its heavy rare earth lines — investors should now look for proof of saleable output. Solvay says it will split dysprosium and terbium at La Rochelle this fall. Carester's Caremag plant at Lacq, France, targets 600 tonnes a year of mixed dysprosium and terbium oxide. But design output is not what rolls off the line in year one. These are near-term claims that will soon face real numbers.

Then come the 2027 claims. Ucore in Louisiana, SRC in Saskatchewan, Iluka at Eneabba in Australia. Each one has a story and a timeline. Energy Fuels pushed its target from late 2026 to Q4 2027 — even after hitting 99.9% purity on a test batch. Good chemistry in a lab and steady throughput in a plant are two different things. The gap between them is where most projects stall.

Lifton's point is blunt. A firm can raise cash, hit a purity mark, and send out a press release. That does not mean it ships tons of oxide, month after month, that a buyer has signed off on. So he now tracks five checks: What was made? How much? What grade? Did a buyer take it? Did they come back for more?

The clock makes this urgent. Xi Jinping lands in Washington on Wednesday, and the Busan truce that suspended new export controls is running out. China sent zero dysprosium or terbium oxide to Japan in the first half of 2026 — but kept sending finished magnets. If the West cannot split its own heavy rare earths at scale, no trade deal changes the math. The shipping test is the only test that counts.

Its "golden tablet" can produce intelligence up to 1,000 times FASTER than regular AI.

And Jeff believes by November 11…

Elon Musk is going to accelerate the demand for this company's patented technology…

Sending shares higher than anyone can imagine. Click here to see the details.

Japan's metals agency JOGMEC signed a deal with Aclara Resources to explore heavy rare earth clay deposits in Brazil. JOGMEC will put up $3 million over three years, with a chance to add $1.5 million in year four. Ionic clays skip the crushing and grinding of hard-rock mining. They are less complex, cheaper, and richer in the heavy rare earths that magnets need. Japan got zero split dysprosium oxide from China this year. Aclara is also building a rare earth plant in Louisiana, with a Q2 2028 start date. This deal is Tokyo building its own supply line, one step at a time.

SMM's weekly wrap on September 19 called the rare earth market "in the doldrums." NdPr oxide slid to 728,000–730,000 yuan per tonne. It held flat early, then fell as sellers cut their asks. Terbium oxide dropped to 6.6–6.65 million yuan per tonne — among the steepest falls in the medium-heavy segment. Dysprosium held at 1.43–1.46 million yuan. Buyers placed small orders and waited. Holmium oxide fell the most of any rare earth this week.

Ucore Rare Metals has pulled in $27 million in grants from the U.S. Army, with a fresh $4.6 million added in September for samarium and gadolinium refining at its Louisiana plant. Samarium-cobalt magnets go into missile systems and jet engines that run too hot for standard neodymium magnets. The Pentagon needs these elements from non-Chinese sources before the DFARS ban takes full effect on January 1. Ucore also closed a C$69 million stock sale in August, giving it cash to push its RapidSX tech toward a first-half 2027 start date. Lifton's test applies here too: the money is in, the Army is watching, but repeat sales have not begun.

A new type of AI called "Accelerated AI" is about to take the world by storm…

And stocks connected to it are already breaking out: 133%… 210%… and even 320% or more just in the last few months.

But it's just getting started…

If you want to learn more about "Accelerated AI" – and get name and ticker of the #1 pick to play this opportunity…

Click here for more details. (No purchase necessary.)

The Rare Earth Mines, Magnets & Motors 2026 event runs September 28–30 at the Westin Harbour Castle in Toronto. It covers the full mine-to-magnet chain. NASA's Reid Wiseman gives the keynote.

— Adamas Intelligence

SMM's Monday morning report said NdPr alloy and terbium both stopped falling and held firm. Dysprosium stayed flat. Magnet buyers are shopping for the best price and waiting, and no one knows yet whether stockpiling ahead of the holiday will kick in.

— Shanghai Metals Market, Sep 21

ChinaFile ran a panel preview of the September 24 Xi-Trump meeting. The Busan truce, signed October 30, 2025, suspended new export controls for one year and reduced some tariffs, but left existing trade curbs in place. Whether it gets renewed is an open question. Each side blames the other for the stall.

— ChinaFile, Sep 17

Carester's plant in Lacq, France, is on track to become Europe's first large-scale rare earth recycling facility when it starts this quarter. Design output: 600 tonnes a year of mixed dysprosium and terbium oxide — about 15% of world supply. USA Rare Earth took a 13.6% stake in Carester in July.

— USA Rare Earth / Mining Technology