As of December 24, 2025, at 21:16, major U.S. stock indices are fluctuating near high levels, with some investment banks raising their target range for the S&P 500 in 2026 to 7,500 points, and optimistic scenarios even pointing to 8,000 points. Capital expenditures for AI infrastructure are estimated by several institutions to potentially accumulate to around $3 trillion, while annual AI-related investment is expected by JPMorgan to rise from about $150 billion to over $500 billion by 2026. As the index targets are raised and the AI investment curve accelerates, the debate over whether we are in an "AI bubble or a super cycle" is rapidly spreading to the cryptocurrency asset market, increasing the risk of mismatched funds and sentiment.
Recently, a set of reports regarding the 2026 outlook from the 15 major investment banks on Wall Street was summarized by certain AI tools in one sentence as "precarious," a highly emotional characterization that quickly spread on social media. Meanwhile, JPMorgan raised its benchmark expectation for the S&P 500 to 7,500 points, even providing an optimistic scenario of 8,000 points under a more accommodative Federal Reserve environment, creating a stark contrast with the "precarious" summary. The investment in AI infrastructure and the macro interest rate path have become the core focus of institutional debate. The real contradiction lies not in whether investment banks are broadly bearish, but in their differing "tolerance ranges" for the combination of AI capital expenditure curves and macro interest rate paths, leading to divergences in the pricing of risk assets, especially AI-related targets and on-chain assets.
Core of the Event
On December 24, 2025, at 08:00 UTC+8, multiple media outlets reported that the market outlook for 2026 from the 15 major investment banks on Wall Street was summarized by AI systems as "precarious." At the same time, JPMorgan raised its benchmark target for the S&P 500 in 2026 to 7,500 points, providing guidance of up to 8,000 points in a more accommodative scenario. According to institutional estimates, AI infrastructure-related capital expenditures are expected to accumulate to around $3 trillion over the next few years, while JPMorgan warned that AI investment could rapidly increase from about $150 billion to over $500 billion by 2026, with the financing structure shifting from corporate free cash flow to more reliance on debt and equity financing. In the context of generally raised index targets, "precarious" is more of an emotional label for the coexistence of high valuations and high Capex, rather than a systematic pessimistic judgment on the economy or corporate profits.
On the other hand, traditional institutions are accelerating their layout in on-chain assets: BlackRock, Fidelity, JPMorgan, and others are continuously taking action in dollar-denominated tokens and the Ethereum ecosystem. Some research institutions predict that the total scale of related tokens could reach hundreds of billions of dollars or even higher by around 2028, while JPMorgan provides a more conservative estimate of around $500 billion. The "Stablecoin Act" passed in July (a market colloquial term that refers to direction rather than the complete text) has driven a significant increase in compliant dollar tokens and Ethereum assets, indicating that institutions have begun to incorporate on-chain assets into their medium- to long-term asset allocation frameworks.
News Aspect: Consensus and Divergence Among Investment Banks
From the publicly available viewpoints, JPMorgan, Goldman Sachs, Morgan Stanley, and others have a certain consensus on the baseline scenario: assuming the Federal Reserve steadily shifts to moderate easing within the next two years, the U.S. economy maintains low to moderate growth, and corporate profits continue to show positive growth driven by AI and digital transformation. Within this framework, JPMorgan dares to raise its S&P 500 target to 7,500 points, with some more optimistic scenarios even projecting up to 8,000 points, while Goldman Sachs and Morgan Stanley are more moderate in their projections but still incorporate "AI-driven profits" and "AI-supported valuations" into their core logic. The true consensus at the news level is that AI has transitioned from a "thematic story" to a core variable embedded in profit forecasts and valuation models, but judgments on its marginal return rates and duration are highly divergent.
Under the AI mainline, most investment banks acknowledge that the next two to three years will be a capital-intensive period, with high-performance computing power, data centers, electricity, and network upgrades consuming a large amount of cash flow. Some institutions suggest that current AI Capex is still less than 20% of the estimated total of $3 trillion, indicating that there is significant room for annual capital expenditures to rise in order to complete the infrastructure layout. This directly drives the upward revision of revenue expectations for tech giants while raising the parameter settings for growth stages in long-term discounted cash flow models.
In contrast, JPMorgan's latest report adopts a noticeably more cautious tone: it estimates that AI-related investment could leap from about $150 billion to over $500 billion by 2026 and issues a risk warning regarding the trend of "shifting from free cash flow to debt financing," expressing concern that the current pace of Capex expansion may outpace the realization of actual profits. Within this framework, AI is difficult to view as a completely risk-free infrastructure revolution; it resembles a "high-stakes game" driven by low interest rate expectations, stock price performance, and social sentiment.
In the cryptocurrency-related segment, major banks have not provided a unified stance, but there are several implicit consensus points: compliant dollar tokens are seen as a key infrastructure bridging traditional finance and the on-chain world, while Ethereum and several Layer 2 solutions are viewed as potential settlement and compliance modules rather than purely speculative assets. The increasing attention of investment bank reports to Ethereum and dollar tokens essentially acknowledges the roles of "on-chain dollars" and "on-chain settlement layers," rather than signaling a simple bullish outlook on all on-chain asset prices.
Financial Aspect: $3 Trillion AI Capex and Asset Allocation Shift
From a financial perspective, the expected $3 trillion level of AI infrastructure Capex is one of the key variables influencing global asset allocation in the coming years. Institutional estimates show that the actual deployment of AI Capex currently accounts for less than 20% of this total, indicating that there are still potential expenditures in the trillions of dollars that will unfold across various segments such as data centers, GPU clusters, power facilities, cooling systems, and network bandwidth. Such a scale of capital expenditure directly changes "who pays, how they pay, and the duration of payment," thereby reshaping the risk premium structure between equities, credit bonds, and high-risk assets.
In the early stages, tech giants primarily relied on their strong free cash flow to cover AI Capex, which appeared relatively safe on paper. However, as the scale of investment accelerates, as pointed out by investment banks and Silicon Valley KOLs, the sources of funding are showing marginal changes: more debt financing, more aggressive equity refinancing, and collaborations with sovereign funds, pension funds, and long-term capital are becoming the norm. Debt financing means an overall increase in corporate leverage; if the realization of profits from AI projects lags behind rising debt costs or declining market risk appetite, it will amplify profit volatility.
Passive funds and index funds have taken on the role of "automatically leveraging" in this round of the AI wave. When the S&P 500 and Nasdaq are passively lifted due to the rising market capitalization of a few AI giants, the funds tracking these indices passively increase their positions in related tech stocks, further pushing up valuation centers. Meanwhile, passive funds are also flowing into the on-chain asset market through ETFs, compliant dollar token products, and Ethereum-related financial products, with the logic being:
First, the chain of "AI Capex → tech stock valuation → risk appetite." When the market expects AI investment to bring higher long-term profits, tech stock valuations are elevated, risk appetite improves, and some marginal funds continue to flow into derivatives and on-chain assets.
Second, the logic of "dollar-denominated tokens/ETFs → on-chain dollar liquidity → passive funds in the crypto market." When the total amount of compliant dollar tokens increases and ETF products expand, on-chain dollar liquidity strengthens, providing more ample market-making and leverage space for DeFi, the Ethereum ecosystem, and some AI narrative tokens.
The essence of the financial aspect is that the capital expenditure and leverage expansion around AI in traditional markets are indirectly providing a pro-cyclical liquidity amplifier for the crypto market through passive funds and on-chain dollar channels.
Sentiment Aspect: The Narrative Pull Between Bubble and Super Cycle
In social media and KOL circles, the emotional divide between AI and the market is more pronounced. In the bullish camp, reports from asset management giants like BlackRock are repeatedly cited: they emphasize that AI is a new infrastructure cycle, arguing that "investment diversification is no longer safe, and opportunities are concentrated in a few AI giants," likening the current phase to the high Capex construction period of early cloud computing and mobile internet. Observers in the crypto space, such as Haotian, directly assert that "the $3 trillion AI Capex has just begun," believing that any view characterizing AI as a bubble underestimates the long-term value of computing power and data infrastructure.
The bears and cautious factions focus on the mismatch between investment speed and return rates. Silicon Valley KOL Wang Chuan pointed out that the current shift of AI Capex from free cash flow to debt financing is a signal that must be taken seriously: as capital expenditure increasingly relies on external financing, the degree of corporate and market bets on future profits is inadvertently heightened. If actual profit data cannot match market expectations around 2027, valuation corrections and deleveraging pressures will arrive simultaneously. JPMorgan's warning about "AI investment rising from $150 billion to $500 billion" is, in fact, a concern about this leveraged curve.
The phenomenon of "the outlook from the 15 major investment banks being summarized by AI as 'precarious'" is itself a typical example of how sentiment can be amplified by algorithms and headlines: it uses an extreme term to cover the full spectrum from cautiously optimistic to aggressively optimistic, reinforcing the market's sensitivity to the bubble narrative.
In the crypto community, this emotional tug-of-war is even more evident. On one end is a strong FOMO regarding the AI super cycle and the "on-chain AI track," promoting a composite narrative of "AI + DeFi + computing power returns"; on the other end is concern over a "dual bubble of AI and crypto," arguing that the current on-chain AI projects lack a solid profit transmission with traditional AI Capex, making them vulnerable to valuation repricing when macro sentiment declines.
Deep Logic: Implicit Assumptions in the Bull-Bear Game
From a deeper logical perspective, the bull-bear game surrounding the AI mainline hinges on three sets of variables: interest rate path, AI profit realization rhythm, and financing structure and leverage levels. The numerical divergence between bulls and bears essentially reflects differing estimates of the "return/risk ratio" after combining these three sets of variables.
First is the interest rate path. If the Federal Reserve executes a relatively smooth rate-cutting path over the next two years and maintains long-term rates within a controllable range, the discount cost of the $3 trillion AI Capex will be relatively manageable, and the pressure on tech stock valuations and index points will be small, making the S&P target of 7,500 or even 8,000 points "calculable" in models. Conversely, if inflation proves stickier than expected and policy shifts more slowly, the net present value of AI projects will significantly decline, financing costs will rise, and the valuation system will need recalibration, increasing the pullback pressure on overvalued AI stocks and related on-chain assets.
Second is the rhythm of AI profit realization. Currently, the most direct beneficiaries of profits are GPU suppliers and a few cloud computing giants, which quickly reflect AI demand in their profit statements through hardware sales and cloud computing rentals. However, broader corporate clients need longer trial and deployment periods to embed AI into production processes, management systems, and even business models, indicating a time lag between "high GPU sales prosperity" and "overall improvement in corporate profits." If by around 2027, there is no marginal improvement in overall corporate profit statements that matches AI Capex, the market will be forced to lower the "long-term profit multiplier for AI."
Third is the financing structure and leverage levels. When AI Capex is primarily funded by corporate free cash flow, the market can accept a certain degree of investment front-loading; however, as the proportion of debt and equity financing increases, repayment pressure and equity dilution will amplify shareholders' sensitivity to return rates. JPMorgan's warning about the $500 billion level of annual AI investment is precisely based on concerns about this "leveraged Capex."
Within this framework, the bull's calculation is roughly: moderate easing + AI profit realization exceeding expectations + controllable leverage → the S&P 500 target range of 7,500-8,000 is supported by models, leading AI leaders and their ecosystems to have premium space, while compliant dollar tokens and the Ethereum ecosystem benefit from risk appetite and liquidity expansion.
The bear calculation is: interest rates declining less than expected + AI profits realizing slower than Capex expansion + debt-driven leverage increase → tech stock valuations correcting, and the crypto market facing pressure during deleveraging, especially AI narrative tokens lacking real profit support are likely to face revaluation.
The essence of the divergence between bulls and bears is not "whether AI has value," but rather "under what combination of interest rates and profit paths is the current valuation and investment pace reasonable."
Crypto Perspective: What Position Are On-Chain Assets In?
From the perspective of the crypto market, Wall Street's AI narrative is reshaping the pricing framework of on-chain assets through three main lines: the battlefield of dollar tokens, the roles of Ethereum and L2 as settlement and compliance modules, and the pseudo-correlation risks of AI narrative tokens and computing power tracks.
In the battlefield of dollar tokens, traditional institutions like BlackRock, JPMorgan, and Fidelity are actively participating through custodianship, bottom-line assets, and compliance frameworks, making this area a core entry point for institutional competition. Research institutions like Bernstein have given optimistic expectations for the future scale of related tokens in the trillions of dollars, while JPMorgan is more cautious, only seeing around $500 billion in potential by around 2028. The core divergence lies in the pace of regulatory implementation and the speed of global payment/settlement adoption of on-chain dollars. These dollar tokens play a role in the crypto market not as a single asset, but as the "monetary base" of the entire on-chain dollar liquidity, with changes in their scale directly affecting leverage and pricing in the Ethereum ecosystem and other public chains.
Ethereum and various L2s are gradually shifting from being viewed as "speculative targets" to "compliance modules and settlement layers" from the institutional perspective. The advancement of the "Stablecoin Act" in July has guided some institutions to passively increase their holdings in Ethereum, partly to support the settlement needs of on-chain dollar tokens and partly to reserve technical and asset foundations for potential future RWA and compliant DeFi businesses. For the AI wave, infrastructure projects related to data, privacy computing, and cross-chain settlement within the Ethereum ecosystem have the opportunity to form a certain degree of functional coupling with traditional AI Capex, but currently, projects that truly achieve revenue sharing remain few.
AI narrative tokens and the computing power track are the most controversial aspect. Many on-chain AI projects are narratively tied to GPU computing power, AI inference services, or data markets, but their revenue models and the transmission paths of traditional AI Capex are not clear: traditional AI investments mainly flow into large GPU suppliers, cloud providers, and leading AI model companies, while on-chain AI projects that cannot directly connect with these funding flows are likely to become "narrative resonance rather than cash flow resonance."
At the current stage, simply viewing on-chain AI tokens as "high Beta assets of AI Capex" carries significant risk, as the revenue and profit elasticity of the vast majority of projects have not formed a verifiable linear relationship with the $3 trillion level of traditional AI investment.
Outlook: Three Scenarios and Key Observational Indicators
The market developments in the coming years can be framed around three scenarios: optimistic, neutral, and risk.
In the optimistic scenario, the Federal Reserve achieves moderate easing, long-term interest rates remain within a market-acceptable range, and AI-related corporate profits are released in line with or slightly exceeding expectations, with the $3 trillion AI Capex seen as a reasonable investment. The S&P 500 converges towards 7,500 points or even 8,000 points, with a few AI leaders obtaining excess returns, and the demand for compliant products in the dollar token and Ethereum ecosystem continues to grow, leading to a structural market in the crypto space under the resonance of improved risk appetite and liquidity expansion. At this time, quality public chains, leading DeFi, and on-chain infrastructure with clear connections to traditional finance have a relatively high advantage.
In the neutral scenario, AI investment remains high, but corporate profit realization shows significant stratification: upstream segments like GPUs and cloud computing are highly prosperous, while the AI adoption progress of most traditional companies is uneven, leading to a market adjustment of the long-term profit multiplier for AI without systemic pessimism. The S&P 500 maintains high-level fluctuations, with index performance driven more by a few weighted stocks. The crypto market shows clear track rotations: compliant dollar tokens, the Ethereum ecosystem, and some projects with real income attract funding, while purely narrative AI tokens and long-tail assets lacking fundamentals face pressure.
In the risk scenario, the market gradually recognizes some AI investments as a temporary bubble: the Federal Reserve's easing is less than expected, or renewed inflation forces interest rates to remain high, while corporate profits around 2027 fail to confirm the high return assumptions of AI Capex. In this combination, tech stock valuations experience systematic corrections, credit spreads widen, and debt-driven AI projects face refinancing pressures. The crypto market often becomes a "deleveraging outlet" in such scenarios: the growth rate of dollar token scales slows or even contracts, Ethereum and other mainstream assets adjust in tandem under reduced liquidity and declining risk appetite, and AI narrative tokens face more severe valuation contractions and liquidity risks.
In practical operations, investors need to pay attention to several key indicators:
First, AI-related Capex guidance and tech stock profit revisions, including capital expenditure plans and profit expectation changes from GPU suppliers, cloud providers, and AI application companies; second, the difference between Federal Reserve policy expectations and actual implementation, especially the combination of long-term interest rates and inflation expectations; third, the total amount of dollar tokens, the intensity of on-chain activity on Ethereum (such as L2 transaction volume, DeFi lock-up), and the advancement pace of various institutional compliance products (ETFs, custodianship, payment solutions). These indicators collectively form a real-time verification tool for whether valuations and liquidity deviate from fundamentals in the "AI + crypto" overlapping cycle.
Strategy and Risk Warning: Creating Distance Between Narrative and Data
For crypto investors, the biggest challenge in the current environment is: how to create sufficient distance from the AI and macro narratives, using data rather than emotional metrics to assess assets. The AI-related stories can indeed bring higher risk appetite and liquidity, but if the projects themselves lack clear revenue paths and cash flow distribution mechanisms, even if the narratives resonate highly with Wall Street's AI reports, it is difficult to sustain valuations.
From an asset allocation perspective, first, be wary of excessive concentration on a single narrative or a few long-tail assets, and moderately diversify into mainstream assets and infrastructure supported by real use cases and on-chain data; second, emphasize liquidity and leverage management, avoiding high-leverage cross-market bets during periods of high macro uncertainty and accelerated AI investment curves; third, for high Beta tracks like "on-chain AI" or "AI computing power coins," adopt more position control and dynamic stop-loss strategies, rather than endorsing short-term volatility with long-term logic.
AI is reshaping Wall Street's valuation framework and capital flows, but crypto assets ultimately need to be priced using on-chain data, revenue, and risk premiums, rather than being simply categorized with phrases like "precarious" or "super cycle."
Join our community to discuss and become stronger together!
Official Telegram community: https://t.me/aicoincn
AiCoin Chinese Twitter: https://x.com/AiCoinzh
OKX benefits group: https://aicoin.com/link/chat?cid=l61eM4owQ
Binance benefits group: https://aicoin.com/link/chat?cid=ynr7d1P6Z
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support@aicoin.com，本平台相关工作人员将会进行核查。
