Baron Technology Fund | Q2 2025

During the second quarter, Baron Technology Fund® (the Fund) posted robust gains, increasing 31.06% (Institutional Shares), outperforming the MSCI ACWI Information Technology Index (the Benchmark), which gained 23.29%, and the broader S&P 500 Index, which rose 10.94%. For the first half of 2025, the Fund increased 11.67%, outperforming both the Benchmark and the S&P 500 Index, which gained 8.93% and 6.20%, respectively.
 

Why Baron Technology Fund Stands Out

Unwavering focus

We manage the Fund with an unwavering focus on delivering superior long-term returns by identifying and investing in a portfolio of technology winners with sustainable competitive advantages and durable growth prospects from technology innovation, advances, and utilization. Transformative technology trends—such as AI; cloud computing; autonomous transportation; robotics; and digital commerce, media, and services—will shape the future, disrupt existing industries, and drive long-term investment returns.

Active management

We believe our disciplined, active management approach has enabled us to differentiate the Fund from and outperform our Benchmark and competitive passive indexes and ETFs, such as the Invesco QQQ ETF.

• Navigating market concentration. Today’s markets have become increasingly concentrated due to the significant stock outperformance and market capitalization gains of the leading technology companies. The top 10 companies in the Invesco QQQ ETF account for 53% of the portfolio, and the Magnificent Seven occupy seven of the top nine slots (Broadcom Inc. and Netflix, Inc. were the other names mixed in with the Magnificent Seven). This concentration poses risks for passive technology investors, as the portfolio weights are rooted in historical returns and current market capitalizations, not future growth prospects or expected returns. We take a different active management approach. We emphasize companies with strong growth opportunities, competitive advantages, and promising returns. While our top 10 holdings, representing just under 54% of the portfolio, include five of the Magnificent Seven and Broadcom, we size positions based on our research-based conviction, not market capitalization. Our top ten investments also include leading innovators such as Spotify Technology S.A., Taiwan Semiconductor Manufacturing Company Limited (TSMC), The Trade Desk, and Zscaler, Inc., which are often overlooked or underrepresented in passive indexes. Our stock selection strategy aims to balance exposure to proven leaders and emerging innovators, managing concentration risk while also targeting sustainable growth.
• Portfolio differentiation. We highlight several other factors that differentiate the Fund from passive competitors. (i) Faster growth. We prioritize businesses with robust and durable revenue growth—our “faster-for-longer” tagline—and profitable, cash-generating business models. On a trailing three-year basis, our portfolio holdings have delivered revenue growth of 29.6% and earnings growth of 40.1%, far outpacing both the Benchmark (9.7% and 9.7%, respectively) and Invesco QQQ ETF (11.4% and 10.6%, respectively). (ii) More small and mid-caps. The heavy mega-cap concentrations in the Benchmark and Invesco QQQ ETF leave little room for more than slight exposure to small- and mid-cap stocks, resulting in passive investors missing out on some of the market’s most compelling growth stories. In contrast, our Baron research heritage and active management approach enables us to provide Fund investors with material investments in unique, higher-growth businesses. The Fund portfolio ended the second quarter with a 38.3% allocation to small- and mid-cap stocks, far above the Benchmark and Invesco QQQ ETF, which have weights of just 8.1% and 11.2%, respectively.

Superior long-term performance

We believe our research-based, active-management approach has enabled the Fund to outperform the Benchmark, S&P 500 Index, Invesco QQQ ETF, and Morningstar Technology Category (the Category) since inception (December 31, 2021) and all trailing periods shown below. In addition, Morningstar has ranked the Fund in the 4th percentile of the Category for the trailing-three-year period and in the 8th percentile since inception.

Performance data quoted represents past performance. Past performance is no guarantee of future results. The indexes are unmanaged. Index performance is not Fund performance. Investors cannot invest directly in an index.

As of 6/30/2025, Morningstar Technology Category consisted of 271, 248, and 239 share classes for the 1-year, 3-year, and since inception (12/31/2021) periods, respectively. Morningstar ranked Baron Technology Fund in the 9th, 4th, and 8th percentiles for the 1-year, 3-year, and since inception periods, respectively.

Morningstar calculates the Morningstar Technology Category Average performance and rankings using its Fractional Weighting methodology. Morningstar rankings are based on total returns and do not include sales charges. Total returns do account for management, administrative, and 12b-1 fees and other costs automatically deducted from fund assets.

Market Backdrop

Last quarter we wrote that the short-term market is a confidence game and subject to swings in investor sentiment. This quarter, with the potent cocktail of the tariff pause and the steady advancement of AI, investor optimism was reinvigorated.

Despite declining by double digits in early April, all major U.S. indexes rebounded to finish in positive territory for the quarter. Risk-off sentiment to start the period centered around President Trump’s April 2 tariff announcement, which Wall Street strategists viewed as more severe than expected and likely to slow economic growth, boost inflation, and increase the odds of a recession. The S&P 500 Index fell more than 12% over the next four trading days, nearly entering bear market territory from its all-time high in February (down almost 19%). The technology-heavy NASDAQ Composite Index officially entered a bear market on this retreat. After bottoming on April 8, U.S. equities rebounded due to positive trade developments. On April 9, President Trump announced a 90-day pause on most reciprocal tariffs to allow for negotiations. China was the only country excluded from the pause, but U.S. and Chinese officials eventually agreed to a de-escalation of tariffs on May 12, alleviating investor concerns and contributing to additional market gains. Other market catalysts included resilient corporate earnings amid tariff-driven pressures, dovish Federal Reserve commentary following muted May inflation, improving consumer sentiment, a ramp in M&A and IPO activity, and AI tailwinds from NVIDIA Corporation’s strong earnings results.

Second Quarter Performance

As stated above, we manage the Fund with a focus on identifying long-term winners benefiting from critical shifts in technology. Transformative technology trends—such as AI; autonomous transportation; robotics; digital commerce, media, health—will shape the future and drive long-term investment returns. For the second quarter, the Fund appreciated 31.1%, beating the benchmark by 777 basis points, largely driven by stock selection.

Our Information Technology (IT) investments led the way, accounting for 72% of the Fund's outperformance in the period. Significant contributors in IT included semiconductor businesses Broadcom, TSMC, and indie Semiconductor, Inc., and systems software companies Cloudflare, Inc. and Zscaler. Our strong stock selection in IT was enhanced by the Fund’s meaningfully lower exposure to Apple Inc., which significantly lagged the Benchmark during the period. Other top relative contributors were a diverse set of digital services leaders, including Spotify, the world’s most popular music and audio streaming service; Duolingo, Inc., a leading language-learning application; as well as two IPOs we participated in during the period: MNTN, Inc., a performance-driven Connected TV advertising company focused on mid-sized advertisers, and Hinge Health, Inc., an innovative digital health care company that automates care for joint and muscle health.

Technology and AI Highlights

Trump Administration

In our last letter, despite heightened uncertainty about tariffs and AI-related trade restrictions, we wrote: “The Trump administration’s words and actions…underscore a commitment to prioritizing AI for national and economic security and a broader goal of maintaining U.S. AI leadership.” On May 7, the administration announced its intention to rescind the Biden administration’s AI Diffusion rule, with a spokesperson saying it would be replaced with “a much simpler rule that unleashes American innovation and ensures American AI dominance.” Earlier in the second quarter, the administration sent mixed signals by banning NVIDIA from selling its H20 low-end AI chip to China, but just this week NVIDIA announced that it had received assurances from the U.S. government that export licenses for its H20 chips will be granted, with the company expecting to resume shipments “very soon.” Also this week, David Sacks, well known from the All-In podcast, and now the White House AI and Crypto Czar, reiterated the administration’s AI policy goals in a CNBC interview: “President Trump assigned us this mission of winning the AI race and there’s really three ways I think we do that. Number one is innovation. We have to outcompete our global adversaries. There is just no substitute for innovation. You can’t regulate your way to winning the AI race. And in the U.S. innovation is done by the private sector... Number two is AI infrastructure. We have to have the biggest, the most, the best AI infrastructure...The third is AI exports. We have to basically export American technology around the world. We have to make it the global standard. In Silicon Valley we understand that the companies that win are the ones that create the biggest ecosystem. They have the most developers on their platform ... [W]e want American technology to be the global standard. We want the most people using it. We want data centers across the world to be adopting the American technology stack…[W]e want to have a mentality of expanding adoption...of American technology. That’s the way we win the AI race.” We agree with these goals, Sacks’ observations, and the criticality of private sector innovation to win the race.

AI adoption and revenue growth

While we are still in the early innings of AI adoption and revenue generation, some notable examples show faster growth than ever experienced before in the digital age. ChatGPT has set records for user acquisition: it hit 100 million monthly active users (MAUs) within just two months of its release (by January 2023), and 100 million weekly (not monthly) active users by August 2023, 200 million by August 2024, 400 million by February 2025, and 800 million by April 2025.

In contrast, it took TikTok nine months and Instagram 2.5 years to reach 100 million MAUs. From an AI monetization perspective, some of the pioneers have achieved stunning growth: OpenAI is projected to do more than $12 billion of revenue in 2025, up from $2.7 billion in 2024; Anthropic’s revenue run rate has quadrupled since the start of this year, ramping from about $1 billion to about $4 billion; and Microsoft reported a total AI revenue run rate of $13 billion for the December 2024 quarter, growing 175%.

AI ROI

Investors and analysts continue to debate AI’s return on investment as companies like the AI frontier model builders (OpenAI, xAI, Anthropic, and Meta) and the hyperscalers (Microsoft Azure, Amazon Web Services, Google Compute, and Oracle Cloud Infrastructure) continue to invest massive amounts in AI training and inference infrastructure. We believe these companies and their leaders have their eyes on the prize–an immense prize. NVIDIA’s CEO Jensen Huang has pronounced: “No technology has ever had the opportunity to address a larger part of the world’s GDP than AI.” If you are not in the race, you are out of this competition and relegated to watching from the sidelines and becoming disrupted and irrelevant, much like Nokia during the smartphone revolution. Today, even 30 years into the Internet digital era, technology spending only accounts for 5%—about $5 trillion—of the $110 trillion of global GDP, with human labor representing about 45% to 50%. The pursuit of Artificial General Intelligence (AGI), which can perform any intellectual task a human can, and Artificial Superintelligence (ASI), which surpasses human intelligence, will disrupt and transform the global economy and “impact every industry” (quoting Jensen again). AI’s leaders understand the prize is in the trillions, not billions.[1] McKinsey has estimated a long-term AI opportunity of $4.4 trillion in added productivity growth,[2] but this strikes us as merely the first fraction of the real 10-to-20-year opportunity. In a recent report Goldman Sachs evaluated the “over $350 billion in cumulative CapEx spent on the first wave of AI infrastructure to develop Large Language Models (LLMs).” The report calculated $925 billion in savings from the first wave across the Fortune 500 by 2030 resulting from headcount reductions and employee productivity gains. Goldman cited benchmarks disclosed by large corporations, including JP Morgan planning to cut 10% of Consumer & Community Banking operations headcount over five years; Amazon seeing a smaller corporate workforce in upcoming years; and IBM citing 30% cost savings in its customer service costs tied to the use of chatbots and agents and 40% to 70% reductions in procurement costs from the adoption of AI tools.

Morgan Stanley Q2 2025 CIO Survey

Highlights we found noteworthy: (i) 37% of CIOs expect IT spending to increase as a percentage of total revenue over the next three years versus 11% who expect it to decline, translating to a 3.4 times up-to-down ratio, which is healthy but below recent highs from the Q3 2023 survey; (ii) AI sustained its position at the top of the CIO priority list, followed by cloud computing and security software; (iii) security software remained the most defensive area of IT spend by a wide margin, followed by compliance software, CRM applications, and AI and process automation; (iv) 60% of CIOs expect to have AI-based workloads in production by the end of 2025; and (v) CIOs estimate 68% of application workloads to reside in the public cloud by the end of 2027 (from 44% today).

xAI’s Grok 4 release

After being founded by Elon Musk just two-and-a-half years ago, on July 9 xAI released its highly anticipated Grok 4 model, built on the largest coherent AI accelerator (or graphics processing units (GPUs)) cluster in the world. Grok 4 benchmarking showed best-in-class results against leading models from OpenAI, Anthropic, Google, and others. These achievements provide more proof that multiple AI scaling laws remain intact. XAI trained Grok 4 on Colossus, the world’s largest compute cluster, with over 200,000 high-performance NVIDIA GPUs all networked together as a single AI training computer. XAI stressed that Grok 4 benefited not only from scaling next-token prediction pre-training to unprecedented levels (100 times more compute than Grok 2), but reinforcement learning (RL)[3] training that refined Grok’s reasoning abilities to new heights (using 10 times more compute than Grok 3). These RL and reasoning capabilities enable Grok to use tools like a code interpreter and web browsing in situations that are usually challenging for LLMs. Grok can use advanced keyword and semantic search tools, run its own queries to find knowledge from across the web, and employ powerful tools to find information from deep within X. On the Grok 4 Livestream, Elon stated that Grok “is the smartest AI in the world;” if given a graduate student GRE exam, “it will get near-perfect results in every discipline of education…from humanities to…languages, math, physics, engineering;” and it “is post graduate PhD level in everything…better than PhD level in every subject…no exceptions.” Elon did allow that Grok “at times may lack common sense and has not yet invented new technologies or discover[ed] new physics, but that is just a matter of time.” As shown below, Grok achieved unprecedented results on a benchmark called Humanity’s Last Exam,[4] a 2,500 question test across a broad range of subjects. The most knowledgeable human might only score up to 5% on this benchmark, but Grok 4 achieved 25% accuracy with no tools and almost 39% with tools; Grok 4 Heavy, which can use multiple AI agents to derive answers, scored over 44% with tools. Elon described the current tools as “fairly primitive,” and said more “powerful tools” would be enabled later this year.[5]

On another exam, ARC-AGI 2,[6] a benchmark that tests a model’s ability to handle reasoning tasks, plotting models based on their score percentage, Grok 4 looks like an outlier, scoring 15.9% (nearly double Anthropic’s Cloud Opus 4 model).

XAI plans to ramp its AI compute cluster towards 1 million GPUs. Elon emphasized that “we’re at the beginning of an immense intelligence explosion…[w]e’re at the intelligence big bang,” but the “real test” for AI is “reality.” Will AI “invent a new technology…improve the design of a car or a rocket or create a new medication…Does the rocket get to orbit? Does the car drive? Does the medicine work?” Elon, who has a history of bold predictions, stated “[Grok] may discover new technologies as soon as later this year and I would be shocked if it has not done so next year…it might discover new physics next year.”

Tesla’s Robotaxi launch. Starting back in late 2016, Tesla equipped every vehicle it produced with a standardized set of cameras, computing hardware, and communication devices. By the end of the second quarter of 2025, Tesla’s fleet had scaled to 7.5 million vehicles, and it had driven over 3.5 billion cumulative miles with Full Self Driving (Supervised). Tesla has been leveraging this data, together with large-scale AI training compute infrastructure and advanced algorithms, to develop its autonomous driving software.[7] After years of investment and development, Tesla launched its Robotaxi service in Austin on June 22. This marked a significant step toward the company’s long-term goal of operating a large, fully autonomous ride-hailing network. As expected, the initial rollout was limited in geographic coverage, user base, and fleet size. In less than a month, however, the company more than doubled the original coverage area in Austin. Additionally, on June 27, just five days after the Robotaxi launch, Tesla completed its first fully autonomous vehicle delivery, transporting a new Model Y from the factory directly to a customer.

AI acqui-hires

There are a limited number of people with strong technical AI backgrounds and experience. This has led to a war for talent among the top companies. Social media has near-daily reports of talent leaving one AI player to join another. Recently, there has been a flurry of aqui-hire deals designed to avoid anti-trust scrutiny. These include: (i) Meta’s $14.8 billion investment in Scale AI, acquiring a 49% minority nonvoting stake and hiring Scale AI’s CEO, Alexander Wang, to lead Meta’s newly formed Superintelligence team; (ii) Google’s $2.4 billion deal to hire Windsurf’s leadership team, beating out OpenAI after its $3 billion acquisition fell apart over Microsoft concerns regarding Windsurf’s intellectual property; (iii) Microsoft’s aqui-hire deal, valued at $650 million, to land the entire leadership team from Inflection AI and appoint Mustafa Suleyman as CEO of Microsoft AI; and (iv) OpenAI’s $6.5 billion of iO Products, a startup co-founded by former Apple design chief Jony Ive, underlying OpenAI’s goal of developing a new class of AI devices.

Below is a partial list of the secular megatrends we focus on:

• Cloud computing
• Software-as-a-service
• AI
• Mobile
• Semiconductors
• Digital media/entertainment
• E-commerce
• Cybersecurity
• Electric vehicles/autonomous driving
• Electronic payments
• Robotics
• Targeted digital advertising

We continue to run a high-conviction portfolio, with an emphasis on the secular megatrends that are enabled by technology broadly. Among others, during the second quarter we initiated positions or increased portfolio weights in the following names:

• Semiconductors and semiconductor equipment: NVIDIA Corporation, Broadcom Inc. and Taiwan Semiconductor Manufacturing Company Limited
• Software: Samsara Inc., Datadog, Inc., and Snowflake Inc.
• Digital advertising: The Trade Desk and MNTN, Inc.
• E-commerce: Eternal Limited
• Digital sports gaming/entertainment: DraftKings Inc.
• Networking equipment: Coherent Corp.
• Digital health care technology: Hinge Health, Inc.

Broadcom Inc. is a leading semiconductor and enterprise software company, generating approximately 60% of revenue from semiconductors and 40% from software. The company is strategically positioned at the intersection of high-performance AI compute and networking infrastructure, while also demonstrating disciplined execution in software. Broadcom has continued its leadership in networking silicon from the cloud era to the AI era and emerged as the most reliable silicon partner for AI foundational model builders to design custom chips to train and inference their frontier models. Shares rose during the quarter on continued momentum in Broadcom’s AI product lines. In its April quarter, Broadcom reported over $15 billion in total revenue, up 20%; over $4.4 billion in AI revenue, up 40%; and over $6.6 billion in software revenue, up 25%. Broadcom continued to demonstrate excellent profitability, with operating margins over 65% and free cash flow margins at 43%. On the company’s earnings call and during other public appearances, Broadcom CEO Hock Tan confirmed that all programs supporting the company’s projected $60 billion to $90 billion serviceable addressable AI market by 2027 were “on track,” inference demand had emerged as an important AI revenue opportunity, and that the company’s AI revenue growth should accelerate to the 50% to 60% level for fiscal years 2025 and 2026.

NVIDIA Corporation is a semiconductor and systems company specializing in compute and networking platforms for accelerated computing. NVIDIA has captured a dominant position in AI infrastructure with a comprehensive portfolio spanning semiconductor accelerators, networking solutions, modular and rack-scale systems, and software. NVIDIA shares rebounded from their first quarter retreat, as signs emerged that the AI cluster buildouts are likely to extend into 2026, with NVIDIA maintaining its leadership. NVDIA delivered a better-than-feared April 2025 quarter, with data center revenues growing 73% to $39.1 billion, despite a $2.5 billion impact from the Trump administration’s H20 China export ban, which would have beaten Street expectations by $5 billion to $6 billion if not for the ban. NVIDIA’s new Blackwell chip accounted for $11 billion in revenue, the fastest ramp in the company’s history. Management provided solid July 2025 quarter guidance, with total revenue expected to grow over $6 billion sequentially adjusted for the China ban. This guidance stripped out all AI-related revenue contributions from China, effectively de-risking that part of its business. Below are some CEO Jensen Huang highlights from the May earnings call:

• "There are now multiple scaling laws. There’s the pre-training scaling law, and that’s going to continue to scale because we have...data that came from reasoning that are now used to do pre-training. And then the second is post-training scaling law using reinforcement learning human feedback, reinforcement learning AI feedback, reinforcement learning verifiable rewards. The amount of computation you use for post-training is actually higher than pre-training...And the third part, this is the part that you mentioned, is test-time compute or reasoning, long thinking, inference scaling...The amount of tokens generated, the amount of inference compute needed is already 100 times more than the...one-shot capabilities of large language models in the beginning.”
• "And the reason that is so different than data centers of the past is because AI factories are directly monetizable through its tokens generated. And so the token throughput of our architecture being so incredibly fast is just incredibly valuable to all of the companies that are building these things for revenue generation reasons and capturing the fast ROIs.”
• "[G]oing forward, I think it’s fairly safe to say that...almost all software will be infused with AI, all software and all services will be...ultimately based on machine learning, and the data flywheel is going to be part of improving software and services, and that the future computers will be accelerated...[We]’re in the beginning of this new era. And then lastly, no technology has ever had the opportunity to address a larger part of the world’s GDP than AI...[T]his is now a software tool that can address a much larger part of the world’s GDP more than any time in history. And so the way we think about growth...has to be in the context of that. And when you take a step back and look at it from that perspective, we’re really just in the beginnings.”

Spotify Technology S.A. is a leading global digital music service, offering on-demand audio streaming through paid premium subscriptions and an ad-supported model. Spotify shares performed well, attributable to both solid underlying results and the company’s durability in an unpredictable macro environment. In the quarter, Spotify continued its path to structurally increase gross margins, aided by their high-margin artist promotions marketplace, podcasts becoming more profitable, and audiobooks coming in at a higher margin than core music. Spotify remains a sticky subscription product with relative resilience in times of consumer uncertainty. Notably, paid users continued to grow at a double-digit pace despite price hikes. Spotify also continues to innovate on the product side, calling 2025 the “year of accelerated execution,” with aims to improve advertising, expand into video, and release a Super Premium tier. We believe Spotify’s market-leading scale, attractive freemium model, product leadership, and pure-play focus will keep the company competitively advantaged over time. We view Spotify as a long-term winner in music streaming with potential to reach 1 billion monthly active users in the next five years.

Apple Inc. is a leading manufacturer of consumer electronics, computer software, and online services. Shares declined amid mounting headwinds, including new U.S. tariffs on Apple’s China-centric supply chain, which are pressuring gross margins, and increased regulatory scrutiny of the App Store model in both the U.S. and Europe. These developments have introduced greater uncertainty around the growth and profitability of Apple’s high-margin services business. While we continue to admire Apple’s brand, ecosystem, and long-term innovation capabilities, the ongoing regulatory overhang and heightened risk to margins and growth prospects led us to reduce the position size and reallocate capital to holdings with more compelling risk/return profiles.

Arista Networks, Inc. is a leading provider of high-performance networking solutions for data centers, cloud providers, and enterprises. Its advanced switching and routing platforms, powered by proprietary software, offer exceptional scalability, automation, and flexibility. Arista generates revenue through hardware sales, bundled software, and post-contract support services, serving major cloud clients like Microsoft and Meta, as well as a growing enterprise customer base. Arista’s stock experienced significant volatility this quarter due to concerns over potential market share losses at Meta. While Arista’s management team has disclosed winning three AI cluster projects at the scale of 100,000 GPUs, uncertainties about the magnitude of its long-term success in AI networking persist. Given this evolving and uncertain landscape, we exited our position during the second quarter to take a short-term tax loss. We will continue to closely monitor developments and may revisit Arista as market dynamics clarify.

Vertiv Holdings Co is a leading provider of critical digital infrastructure solutions for data centers, including power distribution and cooling technologies. We initiated a position late in the first quarter after the stock fell sharply on the market weakness described above. The stock continued to fall in early April on continued negative news flow related to tariffs. We chose to sell our small position (less than 100 basis points) and garner a short-term tax loss. In 20/20 hindsight, this was a mistake, and we should have continued to build our position on weakness. Vertiv rallied back sharply later in the second quarter as the market became more confident around data center capital spending and Vertiv’s sustainable competitive advantage, but the stock was a detractor for the Fund as we did not participate in that rally.

We invest in companies of any market capitalization that we believe will deliver durable growth from the development, advancement, and/or use of technology. We invest principally in U.S. securities but may invest up to 35% in non-U.S. securities.

At the end of the second quarter, the largest market cap holding in the Fund was $3.9 trillion and the smallest was $757 million. The median market cap of the Fund was $30.6 billion and the weighted average market cap was $1.2 trillion.

We had investments in 44 unique companies. Our top 10 positions accounted for 53.8% of net assets.

To end the quarter, the Fund had $100.6 million in net assets. Fund flows were positive in the quarter.

* Individual weights may not sum to the displayed total due to rounding.

We increased our portfolio weights in NVIDIA Corporation and Broadcom Inc. during the quarter. Please see our discussions above for more information.

We added to our position in Amazon.com, Inc. during the quarter to offset dilution because of positive Fund inflows during the period.

During the second quarter, we increased our position in Taiwan Semiconductor Manufacturing Company Limited, capitalizing on temporary stock weakness driven by tariff-related uncertainties. As the unrivaled global leader in semiconductor manufacturing, TSMC continues to benefit from surging demand for its advanced AI chips, alongside easing investor concerns over potential tariffs and macroeconomic risks. We remain highly confident in TSMC’s capacity to achieve sustained double-digit earnings growth over the coming years, supported by its formidable competitive advantages. TSMC’s technological superiority and state-of-the-art manufacturing capabilities cement its leadership in the industry, while its robust pricing power—stemming from its dominant market position and premium offerings—further strengthens its financial resilience.

The Trade Desk is the leading independent internet advertising demand-side platform, enabling agencies to efficiently purchase digital advertising across desktop, mobile, and online video channels. As we addressed in our last letter, we added to our position last quarter after shares declined meaningfully during the quarter following the company’s first earnings miss in 33 quarters as a public company, along with forward guidance that came in slightly below investor expectations. Since this “first miss ever,” we have done substantial research to validate our thesis and continued adding to our position. We believe Trade Desk has been executing well, with the company-wide reorganization complete, a strong pipeline of joint business plans with brands, and the Kokai platform upgrade adoption back on track. While we continue to watch the competitive landscape as Amazon enters the market more meaningfully, we believe Trade Desk represents the best option for advertisers for biddable Connected TV inventory. Trade Desk’s core positioning and strategy has also been validated, as major companies like Netflix, Disney, and Spotify have chosen to open their ad inventory to the company. Long term, we believe Trade Desk has a substantial runway for future growth. The company’s addressable opportunity set remains large and under-penetrated, as it has only an estimated 10% share in the more than $100 billion programmatic advertising market, a small and growing subset of the over $700 billion global advertising market.

We sold Arista Networks, Inc. during the quarter as discussed above.

During the quarter, we divested our position in Intuit Inc. due to concerns that emerging AI-based services would commoditize its core tax preparation offerings, potentially eroding its competitive moat and long-term profitability. We had concerns that advancements in AI-driven tax solutions might challenge Intuit’s market dominance by offering lower-cost or more accessible alternatives. We recognized that Intuit’s shares rebounded later in the quarter, driven by broader market dynamics, strong earnings, and improving investor confidence in the company’s ability to adapt and innovate within its ecosystem. While we remain vigilant about the risks posed by AI disruption, we are closely monitoring Intuit’s strategic responses to these challenges for potential future opportunities.

We sold Vertiv Holdings Co during the quarter as discussed above.

Looking Ahead

Looking ahead, our focus is unwavering: owning category-defining technology businesses that sit at the heart of durable, secular growth trends. Led by visionary, execution-driven management teams, these companies convert breakthrough innovation into expanding free cash flow. We believe this combination uniquely positions the Fund to compound our investors’ capital over the long term.

Sincerely,