"Over 5% 10-Year US Treasury Yield" Fails to Crush AI Investment Frenzy—Is the Real "AI Kill Line" an Inverted Yield Curve?

According to Zhitong Finance APP, as the market's pricing probability for the Federal Reserve to raise rates at least three more times in this cycle significantly rises, the surge in the 2-year US Treasury yield has started to outpace the upward trend of the 10-year US Treasury yield. The US bond market seems to be sending signals that inflation expectations and the term premium associated with fiscal deficits continue to push up the risk-free yields of 10 years and longer, as well as a more complex signal—a rise in yields driven by inflation and fiscal concerns, record-breaking AI financing and bond issuance, but ongoing Fed rate hikes might gradually weaken hopes for the much-anticipated economic “soft landing” and US future economic growth.

Bond market pricing indicates that the US Treasury market is close to issuing a critical signal—a series of Fed rate hikes will begin to shift the market narrative toward the dual risk of a US economic slowdown or even a transition to stock market earnings recession + economic recession.

Last week, the extra yield that investors demanded for holding 10-year US Treasuries over 2-year notes narrowed to just 17 basis points, the smallest gap since early 2025. This so-called yield curve flattening increases the likelihood that the 10-year Treasury yield, dubbed the "anchor of global asset pricing", will soon fall below that of shorter-maturity Treasuries; this phenomenon, where the 2-year US Treasury yield exceeds that of the 10-year, is called a “yield curve inversion”—one of the key leading indicators of a coming recession.

More importantly, the current pricing trajectory in the market is that, as 10-year Treasury yields above 5% have not yet crushed this unprecedented wave of AI infrastructure expansion and expectations of mass AI application adoption, the market is also starting to worry that what could truly end the AI investment frenzy is, in fact, the so-called US yield curve inversion.

The breakout of the 10-year US Treasury yield above 5% is mainly due to increased nominal financing costs, while tech giants (such as Microsoft, Meta, and Google), with their ample free cash flow and high expected return on investment (ROI), can easily absorb interest expenses; by contrast, if the yield curve inverts deeply again or does so due to heightened recession expectations, it usually signals a collapse in financial system liquidity, sharp downgrades in corporate earnings expectations, and a contraction in overall social demand. Even tech giants at that point would be forced to cut capital expenditures on AI data centers and computing hardware, truly touching the “kill line” of the AI boom.

If the curve further inverts in the future, accompanied by tighter credit conditions, corporate financing difficulties, and weaker order expectations, market fears about the end of the AI investment craze may fully escalate: previously, rising discount rates (the "denominator" in the DCF valuation model) suppressed valuations—from a theoretical perspective, the 10-year Treasury yield equates to the critical risk-free rate, or "r," in the DCF model used in equities. Following a “yield curve inversion,” we could also see significant "numerator" cuts in earnings forecasts. Furthermore, yields above 5% on long-term Treasuries—even as yields pull back amid growth concerns—mean that earnings downgrades and widened risk premiums could offset any valuation support offered by lower risk-free rates.

Nevertheless, inversion itself serves as an early warning indicator; it cannot be strictly equated with recession or a peak in benchmark AI and tech stocks. The Federal Reserve Bank of Cleveland has made it clear that the inversion from the end of 2022 to the end of 2024 previously sent unrealized recession signals.

Muse and Astra Expand Application Boundaries: AI Growth Expectations Face Off with Global Funding Cost Pressure

The profit potential brought by AI agents, clashing with valuation pressure caused by the simultaneous rise in both 2-year and 10-year (and longer) US Treasury yields, has become one of Wall Street’s most vivid dichotomies. Yet, as of now, the market’s answer has been: the unprecedented expansion in AI computing power and the AI application frenzy driven by AI agents have outweighed a series of headwinds, including the breakout of 10-year Treasury yields above 5%.

The rise of the 10-year Treasury yield above 5% has yet to destroy the AI rally, but the market is now assessing deeper risks—namely, if the Fed truly begins a sustained monetary tightening cycle, could it not only suppress equity valuations but, under the backdrop of a "yield curve inversion," also transmit into slowing or even declining corporate earnings growth, possibly ushering in a new economic recession? As of the week ending September 25, the Nasdaq 100 Index climbed 3.3%, its largest weekly gain since early August, and even reached a new record high by the close of trading on Tuesday, indicating that AI commercialization and corporate profit expectations continue to prop up risk appetite. However, this resilience only demonstrates that growth expectations have temporarily offset some of the rate pressure, and does not mean tech stocks have escaped the grip of high rates.

The support AI agents provide for the stock market comes from expanding the scope of application and potential sources of revenue. In its Connect summary on September 24, Meta announced plans to integrate Muse with AI glasses in the coming months and to expand its retail, payments, and office connectors, including scenarios like GitHub, Notion, and Box. OpenAI’s GPT-6 Astra, meanwhile, has enhanced computing, software engineering, and multi-step professional tasks. Moving from answering questions toward executing tasks means AI can enter more workflows previously managed by humans, creating new commercial opportunities for subscriptions, usage-based fees, and enterprise services.

This transformation provides critical support for sustainable, long-term profit growth in the AI-driven equity market, and explains why the market remains willing to bet on the AI value chain’s growth prospects even in a higher interest rate environment. Nevertheless, for equity pricing, an increase in workloads must ultimately translate into revenue, profits, and free cash flow to continue offsetting the valuation squeeze caused by rising discount rates.

Wall Street giant Jefferies recently stated that, powered by the dual engines of the AI investment frenzy and better-than-expected profits from AI-related firms, the S&P 500 Index is expected to soar to 8,000 points by the end of 2026 and approach 9,000 points in 2027. Jefferies' core argument is clear and compelling: in a cycle where AI-driven profit growth exceeds the historical average by more than double, fighting the earnings trend is dangerous. Jefferies’ baseline forecast for the S&P 500 at 8,000 points by 2026 is based on earnings per share (EPS) reaching $373 (a year-over-year growth of 35%—well above the market consensus of 29%) and a price-to-earnings ratio of 21.5.

From a reasoning system architecture perspective, a single user command may trigger planning, retrieval, file reading, code execution, tool invocation, results checking, and regeneration. GPUs and other AI accelerators handle model calculations; CPUs handle virtual machines, browsers, task scheduling, and tool execution; long contexts and concurrent sessions expand the need for model state and KV cache management. HBM, server DRAM, and enterprise SSDs serve different storage levels based on performance, capacity, and cost. Nvidia engineering notes explicitly discuss how CPU execution efficiency impacts agent throughput, as well as the necessity of layered cache management between GPU memory, CPU memory, and storage. This suggests that as agent penetration increases, AI applications are likely to deepen across global industries, especially as incremental AI computing demand extends from AI accelerators to high-performance CPUs, storage, and network systems. The ultimate demand scale will depend on the combined effect of user numbers, task frequency, concurrency, and efficiency improvements.

Ultimately, the standoff in the market must be settled by cash flow. AI applications that boost revenue and productivity are likely to raise future cash flows for businesses; higher long-term Treasury yields, meanwhile, increase discount rates and project financing costs, lowering the present value of those same cash flows. For AI data centers in particular, higher capital costs will raise required utilization, pricing, and payback thresholds. Thus, robust demand for AI computing and applications can coexist with valuation pressure on tech stocks: the former determines the growth ceiling, the latter tests the price investors are willing to pay for this growth. If the curve flattens further, possibly accompanied by credit tightening or even yield curve inversion, market focus may expand from valuation multiples to client budgets, order deliveries, cash collections, and broader macro metrics of AI’s ability to drive sustained economic expansion through improved efficiency, forming a more rigorous stress test for the AI super bull market.

From Inflation Trades to Growth Concerns: As Rate Hike Expectations Climb and Recession Alarms Approach, the Bond Market Edges Closer to Raising the Economic Alarm

Last week, the extra yield investors required on holding 10-year US Treasuries over those with a 2-year maturity narrowed to just 17 basis points, the smallest spread since early 2025. This so-called flattening of the yield curve heightens the likelihood that the 10-year Treasury yield soon drops below that of shorter durations. This highly-watched phenomenon is known as a yield curve inversion—hence why the market has revived discussion about curve inversions and their recession warning significance.

It’s necessary to distinguish between the absolute yield level and the spread between maturities: even if 10-year yields are still at highs not seen since 2007, as long as the 2-year yield rises faster, the curve will continue to flatten. Therefore, the 17-basis-point gap seen last week represents the narrowest spread during this period. Currently, the core shift is that the market has become universally vigilant against over-tightening in the future; it cannot yet be called a curve inversion.

Short-term yields are rising, first and foremost reflecting a repricing of the monetary policy trajectory. On September 16, the Fed raised rates 25 basis points, raising the target range to 3.75%–4.00%, while emphasizing robust economic growth and persistently high inflation. Interest rate futures pricing shows traders are bracing for the equivalent of at least three more 25 basis point hikes over the coming year. The 2-year note is especially sensitive to short-term policy changes, while the 10-year and longer maturities reflect expectations for the average path of shorter-term rates as well as the term premium demanded for bearing long-term interest rate risk. Thus, resilience in growth, persistent inflation, and expectations of tighter policy can all push yields up, though not by equal amounts across maturities. Moreover, market pricing is not the same as the Fed actually promising to carry out that many hikes.

Uncertainty in energy supply continues to reinforce this rate transmission mechanism. On September 27, after Donald Trump rejected Iran's ceasefire proposal, he stated he expected talks to resume in the following week; Iranian Foreign Minister Araghchi insisted the reopening of the Strait of Hormuz must come with its conditions met, involving unfreezing assets, lifting blockades and sanctions. There is room for negotiation but normalization of shipping and energy supply remains uncertain. From a macro transmission perspective, persistently high energy costs may both boost inflation—delaying monetary easing—and also erode household purchasing power and corporate profitability, leaving the bond market facing both “higher rates for longer” and “future demand under pressure.”

Historically, an inverted yield curve has provided powerful signals: since the 1960s, an inversion has preceded the past eight US recessions, although this measure did fail to predict the outcome earlier this decade. This is essentially bond investors expressing the view that the Fed has raised rates high enough to impede economic growth. Such an outcome would have broad implications for capital markets, especially stocks trading near historic highs.

As shown above, the US yield curve appears to be flattening, potentially signaling an impending inversion—an occurrence with few historical precedents, one that could foreshadow a slowdown or even a new recession.

This month, the Fed enacted its first rate hike in three years and suggested further increases might come, leading more investors to prepare for such a scenario. This highlights how, after a round of bond sell-offs reflecting mounting price pressures amidst strong growth, a hawkish Fed is shifting the balance among risks.

Zack Griffiths, Head of Investment Grade Bond and Macro Strategy at the research firm CreditSights, noted: “Seeing a 2- and 10-year curve inversion, or significant flattening, makes one question the view that the economy is truly so strong, and that’s now part of what’s priced into the bond market.”

Inversion would reverse the curve normalization seen globally since 2024. Bond investors are typically compensated with higher yields for locking up funds over longer periods and bearing greater uncertainty, hence yield curves usually slope upward.

As recently as last month, the curve was moving in this direction, with long-term yields surging partly on worries that the Fed’s anti-inflation credibility under Chair Kevin Warsh was weakening. But following the September rate hike, shorter-dated bonds led the rise in yields. Traders are betting the coming year will see hikes equivalent to at least three 25-basis-point moves.

Some do not expect an inversion soon, as much of the expected tightening is already priced in, making it harder for short-term yields to rise relative to long-term rates.

Gennadiy Goldberg, US Rates Strategy Head at TD Securities, said: “The market has priced in substantial Fed tightening, flattening the yield curve markedly in recent weeks. This leads us to believe the 2s-10s curve could steepen in coming weeks.”

It is also currently difficult to imagine a sharp economic downturn. According to the latest Bloomberg Intelligence monthly survey, economists have just upgraded their forecasts for US Q3 GDP growth on stronger demand.

But others believe the curve-flattening trend can continue. Ed Al-Hussainy, portfolio manager at Columbia Threadneedle Investments, said that as the Fed tightens policy to cool the economy and inflation, he is positioning for yield curve inversions between the 2- and 10-year, and 5- and 30-year tenors in the coming six months.

He said: “The best sign of monetary tightening at work is a flattening yield curve, and eventually, an inversion.”

As of this week, the 2-year and 10-year Treasury yields were about 4.9% and 5.2%, respectively. The 10-year, a key global fixed income benchmark, is now near its highest levels since 2007.

Inverted curves often reflect concern about growth prospects, as rate hikes are intended to curb loan demand to fight inflation. Slowing growth may eventually create room for the Fed to cut rates, pushing long-term yields below shorter-term ones.

Compiled data shows that since 1978, the 2s-10s yield curve has inverted a median 15 months before a recession, though this interval has ranged from six months to two years.

As seen above, multiple historical cycles show that an inverted yield curve is often a harbinger of recession.

However, in recent years, the curve’s predictive power has come under greater scrutiny. In 2022, several US yield curves inverted, and most economists expected a recession within 12 months. That, however, did not materialize, as the US economy overall weathered the Fed’s 2022–2023 tightening, a regional banking crisis, trade wars, and this year's spike in energy prices.

Although the 2s-10s is the metric most frequently cited by bond investors, policymakers seeking recession signals also examine curves tied to the 3-month borrowing rate. The spread between 3-month and 10-year US Treasury yields remains relatively steep.

The recent flattening of the 2s-10s US Treasury curve has inflicted losses on investors who had positioned early-year trades for a steeper curve. This change is also rippling through US equities—especially bank stocks. Banks, which typically raise funds short-term but lend longer-term, see their net interest margin squeezed as the spread narrows.

The KBW Bank Index, which tracks large bank stocks, entered technical correction territory last week with a 10%+ drop from recent highs.

The shift toward inversion reflects a thorough reset of the economic outlook since the US-Iran war began in February. Before that, traders were betting on a series of rate cuts that would lower short-term yields, not the hikes they are now braced for.

Jamie Patton, Co-Head of Global Rates at TCW Group, said the inversion would be “a signal that the Fed is making a policy mistake.” She said: “It’s tightened too much and will have to cut sharply down the road. So for us, an inverted curve isn’t a sign of macroeconomic health.”