Commentary: Select Strategy

• Growth leadership reversed as first-half winners in semiconductors, equipment and AI infrastructure came under pressure, while software and other prior laggards rebounded.
• The Strategy underperformed as weakness in AI supply-chain holdings and stock selection in communication services, consumer staples and industrials outweighed gains in software and health care.
• Portfolio activity broadened AI exposure toward processors, connectivity and precision timing, while emphasizing businesses positioned to benefit from AI adoption and productivity gains.

Large cap stocks flexed their leadership in a volatile third quarter characterized by sharp rotations among AI-related companies, a spike in bond yields and ongoing conflict in the Middle East. The benchmark Russell 3000 Index gained 1.4% for the quarter with the Russell 3000 Value Index returning 2.3% while the Russell 3000 Growth Index was up 0.5%. In contrast, small and mid cap (SMID) equities endured wide losses with the Russell Midcap Index down 3.0% and the Russell 2500 Index declining 6.4%. Growth stocks in both capitalization ranges trailed their value counterparts by wide margins.

The central macro development was a substantial upward reset in interest rates across the Treasury curve. Escalation of the U.S.-Iran conflict pushed Brent crude above $100 for much of September, reigniting inflation concerns and prompting the Federal Reserve to move from an extended pause back to tightening with a 25 basis point increase. The 10-year Treasury yield rose more than 80 basis points to over 5%, raising both discount rates and financing costs — a particularly difficult combination for longer-duration growth companies and businesses reliant on external capital.

The economy and corporate earnings generally remained robust, although visibility became less clear amid evolving macro conditions. AI remained the dominant market narrative, but investors are becoming far more discriminating about where returns will accrue. Our hypothesis that the trade could begin to shift toward those businesses that benefit from the adoption of AI, both in terms of cost and revenue opportunities, has begun to play out with enthusiasm spreading into areas of health care, industrials, payments and consumer companies that can capitalize on the enormous potential of the technology.

The ClearBridge Select Strategy underperformed its large-cap-heavy Russell 3000 benchmark in the third quarter, with stock selection across communication services, consumer staples and industrials holdings the primary headwinds.

Both growth markets and the Strategy saw a significant reversal in leadership during the quarter. Many of the companies that performed exceptionally well in the first half of the year, particularly semiconductors, semi-cap equipment and AI infrastructure names like Lam Research and Vertiv, subsequently came under pressure. Conversely, some of the biggest first-half laggards, particularly software companies, rebounded meaningfully.

The largest detractors shared a common thread: a reassessment of the pace and durability of AI infrastructure spending that weighed on the AI supply chain regardless of company results. Lam Research retreated from its June high as semiconductor equipment names derated on worries that the biggest AI spenders may have overbuilt, overshadowing record revenue and high gross margins. Vertiv fell on the same sentiment, compounded by second-quarter revenue growth that missed expectations (despite being in the double digits) as large modular projects slipped into the second half of the year; we view the selloff as timing-driven given raised full-year guidance.

AppLovin declined as its e-commerce advertising ramp continued to lag expectations and a delay in mobile-gaming model upgrades left little upside in the core business, though accelerating advertiser signups point to a better setup ahead. Our underweight to Microsoft also detracted as shares rose on faster Azure growth and Copilot adoption, a reminder that software makers monetizing AI are still valuable.

Given the Strategy’s balanced approach to growth, we also participated in the beneficiaries of the July rotation away from AI. Shopify, ServiceNow and other software holdings benefited as investors began reassessing the potential opportunities from AI adoption rather than focusing exclusively on the threat of disruption. Once bucketed as a potential “AI loser,” Shopify became the checkout partner for Meta’s new Muse shopping agent while Amazon blocked the agent, positioning Shopify to be a key infrastructure that agents depend on. ServiceNow delivered a better-than-feared quarter, with contracted revenue growth above guidance and AI contract value crossing $1 billion.

Health care rebounded from depressed valuations amid rotation away from crowded AI positions. Charles River Labs rose as results confirmed the preclinical recovery, with its third straight quarter of book-to-bill above 1x and raised 2026 guidance. Our lack of exposure to Tesla also benefited relative returns as the shares declined on consumer spending concerns.

The Strategy added seven positions during the quarter while exiting 10 others. The largest addition was Advanced Micro Devices, a leading provider of high-performance processors and accelerators across data center, AI, PC, gaming and embedded markets. We believe growing AI and cloud infrastructure investment, continued CPU share gains and expanding GPU opportunities position the company for revenue and earnings upside.

We also added Credo Technology, which provides high-speed connectivity solutions that enable reliable, efficient data transfer across AI data centers and cloud infrastructure. We see a compelling opportunity as the company benefits from leadership in AI scale-out connectivity, expansion into optical solutions, increasing networking complexity and the potential for sustained growth, high gross margins and operating leverage.

New buy SiTime is the leading provider of silicon-based precision timing devices, which are displacing legacy quartz solutions as AI infrastructure, networking and other electronic systems demand greater speed, reliability and power efficiency. Its technological leadership, sticky customer design wins and low share of the broader timing market support a long runway for penetration, while better system utilization can improve the efficiency of power-intensive computing.

While Kratos Defense & Security continues to secure contract awards and remains well positioned across attractive secular growth opportunities in aerospace and defense, it is investing heavily to support future growth and it may take time for conversion of contract wins into meaningful cash flow. This, in combination with the uncertainty around military spending budgets and increased competition, has pressured the stock and led to our exit. We also exited SBA Communications, which operates wireless communications towers, as limited near-term catalysts made the opportunity less attractive relative to alternatives.

The determination of AI winners and losers will take quarters, if not years, to determine. We view the rebound in the recently pressured software industry as a sign of stabilization following a selloff based on fears that generative AI would disintermediate software business models. The reversal was supported by improving sentiment among software makers viewed as enablers of AI adoption and solid earnings results. Software has always been a core exposure in the portfolio, and we maintain confidence in the innovators we own across the industry.

Through private investments, we also have exposure to two of the leading frontier AI model developers, OpenAI and Anthropic, which we believe are among the most innovative companies in the space. More broadly, we are increasingly focused on identifying businesses that can leverage AI to improve productivity, enhance their offerings and drive incremental revenue growth, rather than relying exclusively on companies benefiting from the current infrastructure spending cycle. This is particularly relevant given the inherent cyclicality of hardware spending and the difficulty of predicting when those investment cycles may turn.

We continue to believe the AI opportunity is broadening beyond GPUs and traditional semiconductor exposure. As data centers and GPU clusters become larger and more complex, the need for faster connectivity, more sophisticated networking and greater precision and efficiency is creating substantial new markets for enabling technologies. Recent additions such as Credo and SiTime reflect our efforts to identify differentiated beneficiaries of this next phase of infrastructure development and we believe these trends are materially expanding their addressable markets relative to historical opportunities.

The potential IPOs of Anthropic and OpenAI in late 2026 or 2027 could represent meaningful catalysts for the broader innovation ecosystem. In the meantime, we continue to engage with private companies to deepen our understanding of emerging technologies, competitive dynamics and potential future leaders in AI. These relationships remain an important part of our idea generation process, helping us identify compelling growth opportunities earlier in their development.

The ClearBridge Select Strategy underperformed its Russell 3000 benchmark during the third quarter. On an absolute basis, the Strategy had positive contributions from three of the 11 sectors in which it was invested. The leading contributors were the IT and health care sectors, while the industrials and communication services sectors detracted the most.

On a relative basis, stock selection in IT and financials contributed to performance as did overweights to IT and energy as well as underweights to utilities and consumer discretionary. Conversely, stock selection in communication services, consumer staples, industrials, energy and materials detracted from performance. An overweight to industrials also weighed on results.

On an individual stock basis, the largest contributors to relative returns were Shopify, Charles River Laboratories, DocuSign, ServiceNow and not owning Tesla. The largest detractors were Lam Research, AppLovin, Vertiv, Casey’s General Stores and an underweight to Microsoft.

In addition to the transactions mentioned above, we initiated positions in PNC Financial Services Group in financials and Protagonist Therapeutics in health care. We exited positions in Spotify Technology in communication services, Bloom Energy, Republic Services in industrials, Premium Brands Holdings and Grocery Outlet Holding in consumer staples, GPGI in IT, NRG Energy in utilities and Expedia Group in consumer discretionary.