Around the question of whether "AI is a new round of productivity revolution or the next asset bubble," divisions within Wall Street are rapidly escalating: on one side is Ken Griffin, representing caution regarding current AI valuations, while on the other side is Blackstone President Jon Gray, who recently publicly rebutted the "AI bubble theory," emphasizing that the current demand for AI infrastructure shows a "parabolic" growth. Gray suggests that unlike in the 1990s when companies first crazily laid out telecom networks and then waited for users to come, this time companies are working overtime to build data centers, computing power, and operational systems, trying to catch up to the already existing strong demand. In the market, during the same period, several contracts related to US stocks have seen significant corrections; according to a single source, the USDEX contract fell approximately 16.9% to about 9.9 USDT, while the COHR contract dropped roughly 11.9% to about 207.73 USDT. Coupled with the decline in the U.S. 3-year Treasury yield to about 4.768%, the lowest since September 18, 2024, the juxtaposition of falling interest rates providing a new valuation backdrop and the increased volatility of AI-related assets has amplified the divergent positions on whether "AI is a bubble or a starting point."
Wall Street Split: Gray's Direct Counter to the AI Bubble Theory
At a time of declining interest rates and sharp fluctuations in AI-related contracts, Jon Gray chose to stand on the "not a bubble" side. According to a single source, he stressed that the current demand for AI infrastructure is showing "parabolic" growth, with companies facing already existing and growing demands for computing power and operations, thus having to accelerate the construction of data centers, networks, and supporting systems. Gray deliberately distances this round of AI surge from the telecom bubble of the 1990s—when massive capital was initially invested in expanding capacity, followed by a hope for future traffic to fill the fiber optics; whereas now demand comes first and supply follows, making infrastructure upgrades more of a forced catch-up rather than a blind gamble.
In contrast, billionaire Ken Griffin has previously expressed concerns about whether there is an AI bubble; in the context of rapidly rising valuations and increasing volatility, he represents a more "cautious faction" voice. Gray's high-profile statements have brought to light the previously latent divisions within Wall Street: one side views AI as a new round of infrastructure cycle, trying to justify high valuations with "demand-driven" arguments; the other side benchmarks it against historical asset bubbles, paying more attention to whether ultimate returns can match current prices. For traditional asset investors, this split directly impacts how they understand the risk premiums associated with AI-related stocks and contracts; while in the crypto market, although there is a lack of specific on-chain quantitative data related to AI projects, the public denial of an "AI bubble" by institutions like Blackstone will still psychologically reinforce a portion of investors’ belief in the long-term effectiveness of the "AI narrative," while making another portion of participants who share Griffin's views continue to see AI-related assets as high-beta chips that need constant defense.
From the Telecom Disaster to the AI Frenzy: Demand and Supply Reversed
In Gray's narrative, the true watershed is not about "whether there is a bubble," but rather about who is ahead. The painful lesson from the 1990s telecom industry was that operators first invested huge capital expenditures, laying down fiber optics, switching equipment, and network nodes to the horizon, only to turn back and wait for users and business, resulting in demand climbing less than expected, forcing a large-scale revaluation of assets to "reconcile." In that round, supply had pulled the antenna too high, leaving demand gasping for air halfway up the mountain.
However, in his view, the script today regarding AI is exactly the opposite: companies already sense the genuine pull from the business side towards models, computing power, and automation. The so-called "parabolic" growth refers to the curve of orders, scenarios, and internal applications having raised first, leaving data centers, computing clusters, and operational platforms to passively accelerate their catch-up—Nvidia's launch of the AI factory operation platform, in essence, is also speeding up this infrastructure curve. But Gray does not present this story as a fairy tale; he admits that demand coming first does not mean that capital market pricing must be rational—stock prices and contracts can accumulate optimistic expectations on top of real demand, deviating into trajectories far beyond fundamentals, which is also a key thread that deserves continuous examination in the current valuation debate around AI.
AI Concept Contracts Correcting: Price and Optimism Misalignment
Just as Gray was increasingly confident in narrating the "parabolic demand" story, the market presented a different expression. According to a single source, the USDEX contract has recently dropped about 16.9%, priced at around 9.9 USDT; the COHR contract fell about 11.9%, reporting around 207.73 USDT. These contracts, related to US stocks, are viewed by many participants as a thermometer for market sentiment, and the consistent obvious corrections are hard not to be interpreted as: at a time when verbal narratives are becoming increasingly optimistic, short-term capital has become more conservative in pricing, preferring to apply a discount to positions upfront.
This misalignment is essentially a conflict in the time dimension. Gray is discussing a demand curve that has accumulated over years and is still accelerating, while contract prices reflect expected fluctuations over the coming days and weeks. More importantly, the materials do not disclose the specific components and trading structures of these contracts, nor do they provide quantitative indicators such as trading volume and open interest; hence we can only cautiously view this round of correction as a reflection of changes in sentiment, without being able to prove that "this is already a bubble" or "this is definitely not a bubble." What truly needs to be continuously examined is still the distance between future demand realization trajectories and current prices.
US Treasury Yields Cooling and Nvidia Minerva's Arrival
Beyond emotional fluctuations, a more fundamental context on the same timeline is that interest rates are quietly loosening. According to a single source, the yield on the US 3-year Treasury recently dropped to about 4.768%, the lowest level since September 18, 2024, falling about 13.93 basis points in a day. For all risk assets tied to future cash flows, this decline means lower discount rates—in models, the same growth assumptions yield higher present values. However, the materials did not provide corresponding stock prices or crypto asset reactions; it is difficult to simply interpret this round of yield decline as "assets naturally need to be revalued." It seems more like a slight cooling of the valuation environment rather than the conclusion itself.
Almost simultaneously, the industry provided another set of coordinates: Nvidia launched the AI factory operation platform Minerva, claiming it can analyze billions of data points per hour to detect faults in advance and improve GPU uptime. According to a single source, this platform targets the maintenance and reliability scenarios of AI data centers, behind which is the increasingly dense and expensive GPU clusters demanding higher operational stability. On one side, there is valuation fluctuation caused by a slight decline in the yield curve; on the other side, infrastructure and operational investments like Minerva are accelerating implementation, embodying the "parabolic" real demand Gray described in steel and concrete and chips, drawing a clearer reference line between demand drives and price fluctuations.
When AI Meets Crypto Narrative: Bubble or New Infrastructure
From Gray to Griffin, the debate surrounding AI ultimately points to the same main line: whether this wave described as "parabolic" demand is new generation digital infrastructure or high-position chips pushed up by narratives depends on whether real demand can be continuously realized in the coming years. Gray emphasizes that unlike the telecom era of the 1990s where money was first thrown down to lay cables and then wait for users, today companies are catching up with already existing demands for computing power and operations; whereas Griffin's concerns remind investors that once valuations drift far ahead, even if AI Coin data shows that contracts related to US stocks like USDEX and COHR have corrected, and the U.S. 3-year Treasury yield has once dipped to about 4.768%, it is not enough to prove that risks have been completely released. For crypto assets layered with the AI narrative, their price fluctuations are likely to closely adhere to such macro perspectives and industry progress fluctuations, but lacking specific on-chain addresses, position structures, or protocol-level AI data, it is hard for us to seriously assert that a certain "on-chain signal" has been established, and we can only acknowledge that this remains a story mainly existing at the narrative level. Going forward, the pace of investment in AI infrastructure, the speed of industry application deployment, and the evolution of interest rate paths will together form key observation coordinates in judging whether this AI wave is closer to a bubble or new infrastructure.
Join our community to discuss and grow stronger together!
AiCoin exclusive Hyperliquid benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin exclusive Aster benefits: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到[email protected]，本平台相关工作人员将会进行核查。
