Investors who watched semiconductor stocks tumble over the summer may assume the artificial intelligence trade has already run too far, too fast. Richard Ross — a technical analyst who’s bullish on the U.S. stock market — sees a different setup: Many memory, storage, and semiconductor-equipment stocks remain below their July highs while starting to rise from their 200-day moving averages, a widely watched measure of a stock’s longer-term price trend. For investors who believe the AI buildout still has room to run, Ross’s approach is to favor the companies tied to the physical demands of data centers, including chips, power, and optical networking, rather than wait indefinitely for a cleaner entry point.
Ross, senior managing director and head of technical analysis at Evercore ISI, is also calling for the S&P 500 Index to reach 8,300 this year. That would represent a 7.7% gain from current levels, and a 21% gain for the year. His forecast depends on technology continuing to lead and on the 10-year Treasury yield staying near 5%. (It’s at around 5.1%, the highest since 2007.) His framework offers a useful distinction for investors: A pullback can create an opportunity when a broader trend holds together, but rising interest rates can change the conditions that supported the trend in the first place.
Here is a closer look at Ross’s strategy for approaching a volatile AI trade without treating every technology stock as the same opportunity.
Why a semiconductor pullback doesn’t necessarily mean the AI trade is over
Ross’s bullish view begins with price behavior, or technical analysis, which is the study of market prices and trends rather than evaluating a company’s future earnings. He argues that the S&P 500 had absorbed several apparent obstacles, including a 10-year Treasury yield near 5%, crude oil near $100 a barrel, and an interest-rate increase, while remaining up 16% year to date at the time of publication of this report.
The more contrarian part of his argument concerns the market’s response to bad news. Ross said negative headlines around artificial intelligence coincided with a low point for AI and semiconductor stocks. He does not present that reaction as proof that every AI-linked company will rise. Instead, he treats the ability to hold up after alarming news as evidence that sellers may have already acted.
Related: Nvidia CEO has a blunt message for Americans fearing AI
Ross’s point is deliberately counterintuitive: The most unsettling news can sometimes mark a turning point in a market’s decline. That is a technical observation, not a prediction, that negative headlines are good for stocks. Investors still need to ask whether a company’s business, valuation, and risk tolerance fit their own plan.
Ross’s more practical reason for looking again at semiconductors is that the group’s rebound did not erase the earlier decline. He said many semiconductor memory and storage stocks were still below their highs set in July and were only beginning to rise from their 200-day moving averages. A stock recovering above or holding near that long-term trend line can attract trend-following investors, although the signal can fail and does not guarantee a gain.
How Richard Ross separates semiconductor opportunities from a broad AI bet
A broad bet on AI can conceal very different businesses. Ross named Intel and Dell Technologies as companies in the broader ecosystem, then singled out semiconductor equipment and memory-related names as areas he finds attractive. His examples included Lam Research, Applied Materials, Micron Technology, SanDisk, and Seagate Technology.
The distinction matters because these businesses serve different parts of the technology supply chain. Semiconductor equipment makers provide tools used in chip production. Memory and storage companies are tied to the hardware needed to process and retain data. Ross said that hardware was an underlying component of the AI theme, which is why he doesn’t view the summer decline as evidence that the theme has ended.
“The three most costly words in this business are, ‘I missed it.’ You haven’t missed anything as it pertains to semiconductors, AI, or technology,” Ross said, when asked whether investors who waited for lower prices had missed the semiconductor entry point.
That is a forceful view, and investors should recognize the risk embedded in it. Semiconductors can be highly volatile because the industry is cyclical and because AI expectations can move quickly. Ross himself described the advance of the first half of 2026 as meteoric and the subsequent unwinding of the momentum as painful for many of the same names.
Ross’s preferred way to avoid treating the sector as a single trade is to look for stocks that retained important price floors during the selloff. He pointed to Ciena, an optical-networking company, as an example. Ross said Ciena had risen about 60% year to date at the time of the interview but was also roughly 40% below its 52-week high. His focus was not on the size of either move alone; it was whether the stock held key support, a price level where buying had previously emerged.
A support level is a reference point, not a guarantee. Ross said that comparable AI-related names, including Samsung and SK Hynix, had at times lost roughly half their value during the July and August period while holding support. Investors considering such volatile stocks need to decide in advance how much of a decline they can accept before selling, rather than discovering their limit after a sharp drop.
Related: Zuckerberg, Musk, and Huang take key stand on huge AI issue
Why power and optical networking stocks are part of the AI infrastructure trade
Ross’s AI thesis extends beyond semiconductors. He sees power and optical networking as bottlenecks for building and operating data centers. Data centers require electricity, and AI workloads require large volumes of data to move between computing systems. That creates potential demand for companies serving those constraints, even if the market’s attention is concentrated on the biggest chip names.
He named Ciena and Lumentum as optical-networking examples and Bloom Energy as a power-related name. Ross contrasted those higher-volatility opportunities with NVIDIA, which he characterized as offering more stability and less sensitivity to a stock price surge. The choice is therefore not simply between owning AI and avoiding AI. It is a choice between different risk profiles within the same broad investment theme.
For an investor who wants a more measured exposure, Ross’s comments suggest starting with the question of which bottleneck a company addresses and how much volatility the portfolio can absorb. A company closely linked to a narrow infrastructure constraint may offer greater upside if spending accelerates, but it may also be more vulnerable if expectations change. NVIDIA may have a different risk profile, but Ross did not suggest that a steadier stock is automatically a better fit for every investor.
Related: SK Hynix has a $4 billion problem Intel could help solve
Why Ross doesn’t treat every software rebound as equally strong
Ross isn’t broadly bearish on software. He said the sector had staged a powerful rebound from its lows and argued that AI will affect some software companies more strongly than others. His concern is with application-software companies whose charts have bounced but have not returned to their prior resistance, a price area where rallies have previously stalled.
He contrasted Adobe and Intuit with companies that had recovered more convincingly, including Snowflake, Datadog, Palo Alto Networks, and CrowdStrike. The comparison is about relative price strength, not a declaration that the weaker names are poor businesses. For a technical analyst, the question is whether buyers can push a stock above the price range that capped prior rallies.
Ross’s distinction is useful for investors tempted to buy every company that has fallen sharply. A rebound from a low can be meaningful, but it does not by itself establish a durable uptrend. Investors using charts can watch whether a stock breaks above resistance and holds there, while fundamental investors can pair that observation with their own work on earnings, competitive position, and valuation.
Why the 10-year Treasury yield is the key risk to Ross’s S&P 500 forecast
Ross’s S&P 500 target of 8,300 is tied to which stocks would be leading the advance in the technology sector. He said semiconductors alone accounted for almost 18% of the index at the time of the interview and argued that technology and AI collectively drive more than half of the market. His reasoning is straightforward: An index with that much exposure to technology is unlikely to reach a substantially higher level if technology fails to participate.
The potential spoiler is the 10-year Treasury yield, which influences borrowing costs across the economy and can affect how much investors are willing to pay for growth stocks. Ross called the yield his bellwether. He said a move that held above roughly 5.01% to 5.02% could lead yields to drift higher, increasing pressure on mortgage rates, housing, and risk-taking.
Ross didn’t say that the yield’s brief move above 5% would automatically end his case for a bull run. His concern is a sustained breakout, or a move above a prior ceiling that remains in place. That difference matters. Investors watching the yield should avoid turning a single day’s move into a sweeping conclusion, while still recognizing that persistently higher yields would undermine the easier financial conditions that often support growth-oriented stocks.
More on tech:
How much Oracle stock is Larry Ellison using as loan collateral?
John Ternus’s net worth as Apple’s new CEO
Intel’s secondary share sale explained
How long-term, buy-and-hold investors can use a pullback without waiting forever
Ross rejected the idea that investors should sit entirely in cash until the market’s decline offered a perfect opportunity to buy. His argument is behavioral: People often say they want a pullback, then become more afraid when the pullback arrives because the same bad news that caused the decline feels like a reason to stay out.
“If we get a pullback to trend and the world doesn’t come apart at the seams, then, ‘Yes, that’s where you’re supposed to be putting chips in,’ ” Ross said, on whether a larger market pullback would create a buying opportunity.
Ross’s phrasing contains an important condition: A pullback should occur within an intact trend. Buy-and-hold investors with a long-term view can translate that idea into a disciplined process rather than a demand for a perfect market bottom. They can decide how much equity exposure they want, build positions in stages, and reassess if the market’s broader trend or the investment case for a company changes.
That process also helps separate a diversified portfolio decision from a short-term trading call. Ross favors offense over defense at this stage, but he also sees healthcare as an area that can provide both growth and resilience. He noted that the SPDR S&P Biotech ETF was up nearly 30% year to date at the time of the interview. The figure describes past performance, not a forecast, and biotech can be volatile in its own right.
Ross was less enthusiastic about consumer discretionary stocks because higher oil prices, higher yields, inflation, and geopolitical stress can weigh on consumers. He also said he would not commit new capital to energy after its geopolitical rally, though he said investors already holding energy could view it as a hedge. Those calls reflect his market view rather than as a universal allocation rule.
Related: Sandisk’s stock buyback program explained
The takeaway for S&P 500 investors considering semiconductors
Ross’s central message is that the AI trade should be evaluated by its components, not by a single headline about technology. His picks are semiconductors, semiconductor equipment, memory and storage, power, and optical networking. He is more selective in software, where a sectorwide rebound has not produced equally strong charts for every company.
For buy-and-hold investors who take a long-term view, the decision is simple: first decide whether a volatile AI allocation fits the portfolio, then distinguish between a company with a durable investment case and a stock merely bouncing from a low, and finally, monitor the conditions that could change the broader market backdrop, especially a sustained rise in the 10-year Treasury yield. Ross’s bullish forecast may prove right or wrong, but the framework is more durable than trying to guess the exact day of the next pullback.
Ross’s view is a bullish technical case, not a guarantee of higher prices. Investors considering the companies he named should weigh their own time horizon, diversification needs, and ability to tolerate sharp declines before acting on any chart-based thesis.