On June 5, Oriental Harbor released its report titled "Harbor Perspective | May 2026: Staying Rational Amidst the Frenzy to Go Further," highlighting that the Nasdaq has experienced its largest and fastest increase in the past 20 years—23.7%—over the last two months, primarily driven by the extreme performance of the semiconductor sector: the Philadelphia Semiconductor Index (SOXX) has surged over 73% in the past two months, an unprecedented rise. We can leverage the market's frenzy, but we must not become part of it: 1. In certain segments of semiconductors, extreme product price increases have exacerbated cyclical pressures rather than alleviating them. Continuous price hikes will directly impact downstream profit margins until downstream investment returns become unsustainable, leading to a slowdown in capital expenditure growth. 2. The cash flow of hyperscalers will be depleted by 2026; at the current pace of capital expenditure, the $730 billion in capital spending planned for 2026 will essentially exhaust the cash flow of these four companies. The unsustainable high growth of CAPEX will likely see a downturn in growth rates by 2027, which could significantly affect many companies' linear projections extending to 30 years. 3. The market may be overestimating the speed at which AI benefits will be realized within corporate structures. Even if companies begin to adopt AI on a large scale, due to limitations in the physical world and human organizations, it often takes longer for these benefits to translate into profits. The AI dividend may represent a decade-long productivity narrative, but the market is currently discounting this ten-year story all at once, based on violent stock price increases and capital expenditure growth over just two to three years. 4. The current exuberance in the market is indeed concerning: Jensen Huang's compliment of a "trillion-dollar market cap" led to a 32% overnight increase for a company valued in the hundreds of billions. Over the past two months, it seems no one is concerned about macro risks, nor is anyone discussing the ceiling of models or homogenization; model companies are expected to see their revenues multiply tenfold annually, with valuations soaring into the trillions. CSPs have begun to believe Huang's assertion that "investment in computing power equals revenue," and application scenarios appear to be emerging endlessly. This is not to say that these observations and viewpoints are incorrect, but when the market finds almost no points of contention or concern, sentiment often reaches extremes. It fuels the rise and can exacerbate downturns. As Charlie Munger once said, "Only by being better at refuting our own views than others can we truly deserve our opinions and achieve objectivity and rationality." Do not mistake short-term market sentiment for long-term corporate value. We can take advantage of the frenzy, but we must not become part of it!