Bitcoin price updates: BTC slips back near $75,000 as investors turn elsewhere for gains

Micron surged 21% to top a $1 trillion valuation after UBS raised its price target, while SK Hynix jumped 9.3% in Seoul and Samsung has also crossed $1 trillion this month. The article frames memory chips as a key AI trade, with investors expecting supply shortages and pricing power to persist into 2028. By contrast, bitcoin fell 1.5% to $75,800 as capital and attention continue rotating toward AI-related equities.

Analysis

The key second-order effect is that memory is transitioning from a cyclical commodity to a capacity-constrained AI infrastructure toll road. If pricing power really extends into 2028, the market is effectively underwriting a multi-year capex supercycle, which should pull forward orders not just for MU and Asian peers but for equipment, substrates, advanced packaging, and high-bandwidth memory ecosystems. That favors suppliers with leverage to bit demand growth and hurts downstream hardware OEMs and cloud buyers that cannot pass costs through as easily.

The move is powerful, but consensus may be extrapolating a clean demand curve through a very lumpy supply chain. Memory has a history of overshooting on both sides: when lead times lengthen, customers double-order, inventories build, and the trade can unwind fast once procurement teams see even modest easing in spot pricing. The main near-term catalyst to watch is whether AI server demand remains broad enough to absorb incremental DRAM/HBM capacity without a mix shift or pricing reset over the next 1-2 quarters.

Crypto weakness matters less as a direct substitution trade and more as a signal of where marginal risk capital is flowing. The relative performance gap tells us the market is rewarding identifiable cash-flow duration in AI over narrative-driven assets, which can reinforce momentum in semiconductor names until positioning becomes crowded. That said, when sentiment gets this one-sided, the vulnerability is not a thesis break but a drawdown from any guide-down, export-control headline, or inventory warning.

NDAQ being flat is itself informative: this is not a broad beta rally, it is a narrow factor rotation into AI hardware scarcity. That argues for a more selective expression than simply chasing the group, because the next leg should be won by the names with the most convex exposure to memory pricing and the least customer concentration risk.

Key Decisions for Investors

• Stay long MU for 4-8 weeks, but use call spreads rather than outright stock to reduce gap risk after the recent vertical move; upside remains if channel checks confirm tight HBM/DRAM supply, but a 10-15% pullback is plausible on any inventory normalization headline.
• Initiate a relative value long MU / short NDAQ pair for 1-3 months: the trade captures continued AI hardware scarcity while hedging broad-market multiple compression; keep the short leg small since NDAQ is neutral here, but it helps finance the long convexity.
• Add a basket long in memory-enablers on any 3-5% intraday weakness over the next 2 weeks, focusing on equipment and advanced packaging exposure; the best risk/reward is in names where order books lag the memory price move by a quarter or two.
• Trim or hedge any underweight in crypto-linked risk assets for the next 1-2 months; sentiment is fragile and flows are leaving the space, so BTC-adjacent names may underperform even if Bitcoin stabilizes.
• Set a downside alert on MU if management commentary or channel data implies pricing peaking before year-end; if that happens, expect the multiple expansion to compress quickly and consider taking half profits into strength.