The seven giants’ “valuation cutting” is nearing its end! Muse and Astra ignite “AI FOMO trading,” tech stocks are ready for a major comeback.

According to Zhitong Finance APP, after the Nasdaq 100 Index hit a new all-time high last week, Wall Street giant JP Morgan has joined other leading institutions such as Goldman Sachs, Jefferies, and Yardeni Research in turning bullish on U.S. tech stocks. This has prompted both institutional and retail investors to focus even more on buying the dip strategies during Monday’s U.S. stock market pullback. JP Morgan believes that the overall valuation adjustment for the seven major U.S. tech giants (the “Magnificent Seven” or Mag 7), which dominate the U.S. stock market by weight, may largely be complete, and earnings growth is expected to once again become the main driver of share prices. JP Morgan noted that the Mag 7’s expected price-to-earnings ratio for the next 12 months relative to the broad market has fallen to about one standard deviation below the historical median, hovering at a ten-year low.

At Monday’s U.S. stock market close on September 28, prices of most popular AI computing-related stocks, including AMD, Micron, and SanDisk, came under heavy pressure as oil prices and U.S. Treasury yields rose. However, Nvidia (NVDA.US), the “AI chip king,” saw its stock price rise against the trend after announcing a record $150 billion share buyback authorization. JP Morgan’s optimistic view does not overlook changes in business models: rising AI capital expenditures have increased capital intensity, leading to higher financing needs and pressure on free cash flow, which does justify a certain degree of valuation discount. Nevertheless, the current market has already absorbed a significant amount of these adjustments, and future profit growth could still more than offset the drag caused by further valuation compression.

Apple, Microsoft, Google parent Alphabet, Amazon, Meta, Nvidia, and Tesla are not only key components of market-cap weighted U.S. equity indices, but also shape global AI investment expectations through their reach into AI chips/semiconductors, cloud computing, AI applications, and broad end-market deployments. Previously, an estimated $30 trillion expansion of the S&P 500’s market capitalization over three years through 2025 was largely driven by the Magnificent Seven and the broader AI computing infrastructure supply chain. Therefore, changes in the profits and valuations of these giants simultaneously impact benchmark index performance, global stock market risk appetite, and growth prospects for the AI computing “value chain.”

As Meta Muse and OpenAI's GPT-6 Astra trigger a sweeping wave of AI agents, “AI FOMO trading” (where investors rush to buy or refill their holdings for fear of missing out on AI asset rallies) has made a strong comeback. In terms of the underlying infrastructure of AI computing, Muse and Astra are expected to further broaden the use cases for AI agents. The expansive workload for AI inference that they drive not only increases demand for model reasoning, tool execution, and state management but also intensifies competition among some software companies. Accelerators such as GPUs and TPUs handle model computation; CPUs run browsers, virtual machines, product retrievals, database queries, and transaction orchestration; HBM and server DRAM hold model data, context, and concurrent work environments; enterprise SSDs store product indexes, job records, and persistent states, while high-speed networks and optical interconnects support distributed data exchange.

The so-called "Magnificent Seven" (Mag 7), which account for more than 40% of total weight in the S&P 500 and Nasdaq 100 indexes, are the main driving force behind the S&P 500’s repeated record highs. Top Wall Street investment institutions regard them as the group most capable of generating substantial returns for investors during the biggest wave of technological change since the internet era.

JP Morgan: The Magnificent Seven’s Valuation Reset Largely Complete

According to the latest research report from JP Morgan’s equity strategy team led by Mislav Matejka, the Mag 7 have undergone significant valuation restatements, with their expected price-to-earnings ratio for the next 12 months relative to the market now approaching one standard deviation below the historical median, sitting at a decade low.

JP Morgan’s strategists believe that with the crowded positioning in U.S. tech stocks easing, robust earnings, and more reasonable valuations, the tech sector should regain some of the lost momentum since midyear. They therefore recommend investors re-enter the sector during market pullbacks.

Tech stocks have still led the S&P 500 by a wide margin this year, but the rally has lost steam in recent months amid concerns that massive spending on AI might not deliver the hopeful returns that investors have assumed. Within the sector, the Magnificent Seven’s valuations are at their lowest in a decade, while semiconductor stocks are slowly emerging from a tough period—Anthropic’s Dario Amodei and OpenAI’s Sam Altman have both called for a coordinated slowdown in advanced AI development, which has deepened the sector’s woes.

"We suspect there ultimately won’t be a significant deceleration, because this race remains existential—a winner-take-all contest," JP Morgan’s equity strategists led by Mislav Matejka wrote. JP Morgan believes that although the gains seen in the first half of the year are unlikely to be repeated, major opportunities still exist.

The current low valuations of the Magnificent Seven are not only JP Morgan’s view. According to data from Morgan Stanley Wealth Management Global Investment Committee, the Magnificent Seven’s valuation premium relative to the other 493 stocks in the S&P 500 currently stands at just 10%—the lowest in more than a decade—even as these seven giants maintain an average annual earnings growth advantage of around 45%.

Lisa Shalett, Chief Investment Officer at Morgan Stanley Wealth Management, wrote in the report: “By comparison, we believe these mega-scale cloud giants now appear downright cheap.”

JP Morgan strategists add that the Mag 7, along with the rest of the tech sector, have already experienced valuation compression, with most of the adjustment already behind us. The bank had noted as early as March that the pullback in valuations might have gone too far.

The strategists further point out that changes in these companies’ business models justify the compression in valuation multiples to some extent, including sharply rising AI-related capital expenditures, increased leverage, and decreasing free cash flow. However, they note that much of the valuation reset has already occurred, and the mega-cap cloud companies’ exceptionally strong earnings may continue to support share prices—although some of this effect may be offset by further valuation adjustments.

From “Killing Valuations” to “Earnings Take Over”: JP Morgan and Goldman Sachs See New Tech Sector Rally Catalysts

The opportunity identified by JP Morgan’s strategists comes from a combination of less crowded positioning, lower valuations, and earnings resilience, with a particular preference for semiconductors. According to their latest research, since June, 12-month forward earnings estimates for the semiconductor sector have been raised by about 30%, in contrast to unimproved expectations for software earnings. Their outlook for capex by hyperscale cloud companies is: about $950 billion in 2026, $1.4 trillion in 2027, and $3 trillion in 2030.

On this basis, JP Morgan favors a relative trade of going long semiconductors and selectively shorting certain high-momentum software stocks—while noting that software valuations have already adjusted sharply, making broad shorts inappropriate.

Muse and Astra bolster the practical foundation for this “earnings-led” AI-driven profit surge: agents turn a single user request into a multi-stage workflow of retrieval, planning, tool invocation, code execution, and result validation, with demand for CPU scheduling, memory capacity, storage access, and optical interconnect data transmission all rising together with GPU computation. Muse is already equipped with proprietary cloud-based virtual machines, browser operation, background persistent tasks, and memory mechanisms, while Astra enhances computer operation, software usage, and multi-step professional task execution ability.

More important for investment strategy is whether the massive expansion in AI complex workloads can outweigh falling prices per inference and translate into cloud service revenue, chip orders, and sustainable profits. The market has already responded to expectations for broader adoption: according to Nasdaq’s official weekly report, the strong early showing from Muse and the AGI boom ignited by Astra have significantly fueled AI FOMO trading. On September 22, the Nasdaq 100 Index reached a new all-time high, posting a weekly gain of around 3%.

Goldman Sachs strategists offer an alternative support pathway. Goldman senior strategist and partner Mark Wilson believes the year-end rally doesn’t necessarily have to wait for the U.S. midterm elections to conclude: if inflation cools, growth slows but avoids recession, then the additional tightening expected by the market may not fully materialize. Goldman economist Jan Hatzius notes upside risks to growth have diminished, while equity strategist Ben Snider argues some sectors are enjoying excess profits but, overall, no profit bubble has yet formed. The key for the “Goldilocks scenario” recognized by Goldman’s strategists is that as inflation and interest rate pressures ease, core earnings can remain resilient, a point that underlies Goldman’s conditional optimism.

The AI bull market seems to be seeking the “earnings handover window”—when valuation compression slows, steady earnings per share growth can drive stock prices higher without needing multiples to re-rate; if rate pressures subsequently ease, stable valuations may further add to returns. JP Morgan values the synergy of positive earnings trends and positioning flows, while Goldman’s focus is on improved macro conditions—both highlighting the crucial importance of strong execution in AI-driven tech company earnings.

Wall Street giant Jefferies recently stated that, powered by the dual engines of the AI investment frenzy and AI company earnings beating expectations, the S&P 500 Index is forecast to soar to 8,000 by the end of 2026, reaching 9,000 in 2027. Jefferies’ logic is straightforward and compelling: in a cycle of AI-driven profit growth more than double the historical average, betting against the earnings trend is risky. Jefferies’ baseline forecast for the S&P 500 at 8,000 in 2026 is based on EPS of $373 (a 35% year-on-year increase, far above market consensus of 29%) and a price-to-earnings ratio of 21.5 times.