AL (ArchLoot) fluctuates 117.9% in 24 hours: low liquidity trading triggers drastic price volatility

Analyst Ben Bajarin believes that the demand for AI computing power will continue to grow, and 2027 will be the "bottleneck" year with the most severe industry-wide supply constraints. In 2028, supply and demand are expected to remain slightly tight but balanced. Jay Goldberg, on the other hand, warns that the historical cycle's iron law suggests that massive production capacity from companies like TSMC could lead to a collapse in pricing power and a reversion of profit margins to the mean. Both agree that the era of "blindly buying into the AI supply chain and seeing instant surges" has ended, and investors must now carefully select core assets with resilience in cross-cycle pricing.

Japan's government bond yields have surpassed 3% for the first time in 30 years, triggering global alarm. Nomura states that the epicenter of this global rise in long-term interest rates is within Japan itself—driven by uncontrolled fiscal expansion and expectations of central bank rate hikes, making fiscal risk premiums the main driving force. What’s more concerning to the market is that the continuous rise in Japanese bond yields not only threatens the balance sheets of global financial institutions, but may also burst the AI tech stock bubble and trigger a sudden economic slowdown.