Myuex

Long term energy storage solutions

The thesis that long-duration energy storage is needed to stabilize the grid against surging AI-driven electricity demand is structurally sound and well-supported by the current regime. The worldview shows a collision of forces: AI data center power demand is projected to double by 2030, the grid faces severe connection bottlenecks, and renewable intermittency is a live constraint. The Fed's rate hike cycle is a headwind for capital-intensive storage projects, but the sheer scale of the AI energy demand force—validated by multiple news events showing grid strain, solar curtailment, and battery giants pivoting to data center storage—overwhelms the rate sensitivity for this basket. The strongest upside driver is the confirmed grid bottleneck: PJM's emergency request for backup generators and the finding that half of US AI data centers may never be built due to grid limits directly validate the need for storage. The most probable risk is that gas turbines and nuclear revival compete for the same demand, but the storage thesis is not binary—it benefits from any scenario where grid expansion lags demand growth. The basket's holdings are well-positioned: GWH, EOSE, NRGV, FLNC, and IESVF all produce dedicated long-duration storage hardware, while BEP's pumped storage and renewable generation give it a partial but real exposure. The author's claim that renewables are intermittent is supported by evidence; the claim that long-duration storage can stabilize the grid is speculative but coherent and gaining credibility as grid strain events multiply. The basket's -30.8% realized return since posting reflects the market's prior skepticism and the rate hike shock, but the structural demand thesis has only strengthened since then, making the current depressed valuations a potential entry point for the thesis horizon. The key driver holding conviction up is the AI energy demand force, specifically the confirmed grid connection bottleneck and the BloombergNEF forecast of data centers using 20% of US electricity by 2035. The key driver holding conviction down is the Fed's rate hike cycle, which raises the cost of capital for these pre-revenue companies and competes with the storage narrative for investor attention. The regime is hostile to growth stocks but the structural demand force is powerful enough to justify conviction above the mechanical anchor, which is balanced because the scenario set includes both upside and downside cases with similar weight.

GaN is the future of semiconductors

The thesis rests on a structural physical advantage of GaN over silicon for power efficiency, which is well-supported by physics and confirmed by design wins at Navitas. The strongest near-term evidence is the series of events validating the non-Chinese gallium supply chain: Alcoa's Western Australia gallium FID, Metlen's first European gallium contract, and 5N Plus's US defense gallium arsenide role all directly build the pipeline the thesis requires. However, the regime is hostile to the thesis's timeline: the Fed is hiking into reaccelerating inflation, the 10-year yield is above 5%, and the consumer and finance sectors are deeply bearish, which raises discount rates on NVTS's distant cash flows and may slow enterprise procurement cycles. The AI infrastructure boom is real and supports long-term GaN demand, but the enterprise procurement cycle pressure is confirmed by the Fed's hawkish stance and multiple news stories about delayed IT budgets. The basket's holdings are correctly linked: NVTS is the pure-play GaN chipmaker, FPLSF refines the high-purity gallium, and AA extracts the raw gallium as a bauxite byproduct. The key tension is between the strong structural supply-chain buildout and the hostile macro environment that pushes adoption further out. The upside is concentrated in the supply-chain diversification story, which is accelerating with real contracts, while the downside is concentrated in NVTS's valuation and timeline risk. The mechanical anchor reflects this balance, and I sit slightly below it because the hostile regime and slow procurement cycle are live headwinds that the supply-chain events do not fully offset for the thesis horizon. The strongest upside driver is the non-Chinese gallium pipeline buildout, and the strongest risk is the Fed's hiking cycle delaying enterprise adoption of GaN.