$500 billion is the current valuation that OpenAI has been hyped to reach.
Some have calculated that, in order for this round of AI investment to achieve even a 10% return, humanity would need to allocate approximately $650 billion annually, indefinitely. To put it bluntly, every iPhone user would have to pay an additional $34.72 per month on a long-term basis just to cover the costs incurred by this AI boom.
Morgan Stanley added another layer of complexity: while many believe the bottleneck for AI lies in models and chips, the biggest issue may actually be electricity. By 2028, the United States alone could face a power shortage of up to 44 gigawatts solely for data centers. Addressing this shortfall would require around $2.5 trillion in investments for grid and power infrastructure, plus another $2 trillion for the data centers themselves. The total bill for the entire AI infrastructure could fill multiple pages of A4 paper.
Some astute individuals have now set their sights on power plants, substations, and natural gas pipelines. To their astonishment, this wave of AI enthusiasm appears to be a model competition on the surface, but underneath it represents an unprecedented reallocation of resources. Capital, land, electricity, copper, and chips have all been brought into the same coordinate system, forcibly compressed into a unified settlement within the next decade.
Three things are happening simultaneously.
The first is money. McKinsey estimat...

Against the backdrop of the ongoing rise of AI narratives, Alibaba has delivered impressive results. In contrast to Meituan, which has nearly depleted its profits in the food delivery battle, Alibaba's performance has been commendable. In Q1, revenue reached 247.6 billion yuan, a year-on-year increase of 2%, with net profit at 42.4 billion yuan, a significant increase of 76% year-on-year; on a non-GAAP basis, net profit was 33.5 billion yuan, a year-on-year decrease of 18%, primarily affected by the food delivery competition. Compared to Meituan's dire situation, Alibaba's slight decline is not a major concern. Earnings per ADS were 14.7 yuan, overall meeting market expectations.
By business segment, instant retail revenue grew by 12% year-on-year, Alibaba Cloud increased by 26%, and AI-related businesses have maintained triple-digit growth for eight consecutive quarters. It is worth noting that among domestic technology companies, Alibaba has the most thorough commercialization of AI. The financial structure has also seen some new changes: capital expenditure surged from 11.9 billion yuan to 38.7 billion yuan year-on-year, while free cash flow turned into a net outflow of 18.8 billion yuan, compared to a net inflow of 17.4 billion yuan in the same period last year. This is mainly due to two areas of investment—cloud infrastructure expansion and increased subsidies for food delivery services. Despite cash flow pressure, Alibaba still executed a buyback of 815 million USD, repurchasing 7 million ADS.
More significantly, a surprise announcement during the conference call revealed that Alibaba has developed its own AI chip. There are three highlights: first, it fills the gap left by NVIDIA's H20 in the local market; second, production does not rely on Taiwan Semiconductor, but is outsourced to domestic wafer factories (the market widely speculates it is SMIC); third, ...

it's hot 🔥 til it's not!!