$MYSTMYST Harga

On August 20, the U.S. Treasury recently alleviated pressure on the Treasury market by expanding its long-term bond repurchase program, but a wave of corporate debt financing driven by artificial intelligence infrastructure development is heating up. As investments in data centers, high-end chips, and AI services continue to grow, tech giants like Microsoft, Google, Amazon, Meta, and Oracle are increasing their bond financing efforts. The market expects that the issuance of U.S. investment-grade corporate bonds will peak after Labor Day in September, potentially reaching $200 billion. Data shows that since 2026, the issuance of U.S. investment-grade corporate bonds has increased by 38% year-on-year, with the total issuance expected to reach a record $2.1 trillion for the year, much of which is related to AI capital expenditures. In recent years, tech giants primarily relied on cash flow to support their AI initiatives, but as competition in the industry intensifies, the long-term capital needs for data centers, power, and computing equipment are rapidly expanding, leading companies to increasingly depend on the bond market for financing. The market's focus has shifted from 'Can AI generate profits?' to 'Can substantial infrastructure investments yield sufficient returns?' Some investors are concerned that the expansion of AI debt is altering the allocation of funds in the fixed income market, potentially creating new competition for funding between tech corporate bonds and U.S. Treasuries. Andrzej Skiba, head of fixed income at RBC Global Asset Management, stated that the current supply of AI-related bonds is 'approaching the limit that won't disrupt the market.' Analysts point out that if future AI revenue growth cannot cover the massive investments in data centers and chip procurement, some capital expenditures may face the risk of inadequate returns. The market is also beginning to compare the current AI financing boom with the internet bubble of the early 2000s, cautioning that the speed of capital investment may outpace the speed of business model realization. Although the Treasury's repurchase program helps improve liquidity in the Treasury market, it cannot change the trend of simultaneous growth in government debt and corporate financing needs. The large-scale corporate bond issuance in September may become a new stress test for the Treasury market.

On August 20, the U.S. Treasury recently alleviated pressure in the Treasury market by expanding its long-term bond repurchase program, but a wave of corporate debt financing driven by artificial intelligence infrastructure development is gaining momentum. As investments in data centers, high-end chips, and AI services continue to grow, tech giants like Microsoft, Google, Amazon, Meta, and Oracle are ramping up their bond financing efforts. The market expects that U.S. investment-grade corporate bond issuance will peak after Labor Day in September, potentially reaching $200 billion. Data shows that since 2026, the issuance of U.S. investment-grade corporate bonds has increased by 38% year-on-year, with the total issuance expected to reach a record $2.1 trillion for the year, much of which is related to AI capital expenditures. In recent years, tech giants have primarily relied on cash flow to support their AI initiatives, but as competition in the industry intensifies, the long-term capital needs for data centers, power, and computing equipment are rapidly expanding, prompting companies to increasingly rely on the bond market for financing. The market's focus has shifted from 'Can AI generate profits?' to 'Can massive infrastructure investments yield sufficient returns?' Some investors are concerned that the expansion of AI debt is altering the allocation of funds in the fixed-income market, potentially creating new competition for funding between tech corporate bonds and U.S. Treasuries. Andrzej Skiba, head of fixed income at RBC Global Asset Management, stated that the current supply of AI-related bonds is 'approaching the limit that would not disrupt the market.' Analysts point out that if future AI revenue growth cannot cover the massive investments in data centers and chip procurement, some capital expenditures may face the risk of inadequate returns. The market is also beginning to compare the current AI financing boom to the internet bubble of the early 2000s, cautioning that the speed of capital investment may outpace the realization of business models. Although the Treasury's repurchase program helps improve liquidity in the Treasury market, it cannot change the trend of simultaneous growth in government debt and corporate financing needs. The large-scale corporate bond issuance in September may become a new stress test for the Treasury market.

On the 20th, at the 2026 World Robot Conference, Yushutech CEO Wang Xingxing stated that in 80% of unfamiliar scenarios, robots can successfully complete about 80% of tasks through voice or text commands, marking the arrival of the 'ChatGPT moment' for embodied intelligence. However, the most critical challenge in achieving this moment is the alignment of AI model inputs and outputs with real robots. He noted that after the model processes inputs and outputs, there may be deviations in the robot's operations. Yushutech is developing 'Physical AI Robot Self-Evolution V1.0', which enables AI models to autonomously write control code for robots, followed by training and validation in a simulation environment before being deployed in real machines for feedback analysis. He emphasized that the stronger the foundational model, the greater the self-evolution capability; the more extensive the multi-source data, the more pronounced the scale effects of self-evolution; and the more real machine deployments, the faster the data accumulation and capability iteration, leading to an expanded skill set and a scalable skill ecosystem. (Source: Science and Technology Innovation Board Daily)