As with any investment, your capital is at risk.

“Perseverance is the scarce resource in China, not intelligence. The competitive dynamics are too bloody… you have to do the dirty work.”

Ask a large language model (LLM) about the barriers to building a successful business in China, and it will not tell you this. Instead, that observation came from Dr Kai Yu, founder and chief executive officer (CEO) of Horizon Robotics, over some lunchtime squirrel fish, on a warm Beijing afternoon during one of our investment trips.

The exchange was a reminder that, as artificial intelligence (AI) accelerates the availability and synthesis of vast troves of information, an active investor’s edge will come from doing something that the LLMs cannot: having quiet, unscripted and unrecorded conversations with each other, human to human.

Our access to company leaders presents an opportunity for them to speak openly with us about questions relevant to our long-term investment horizon. Learning what has been omitted from slides, scripts and consensus narratives now matters more than ever. And Baillie Gifford has been honing that skill for more than a century.

Having such conversations with companies is the bread and butter of what we do in the Long Term Global Growth Team. In the second half of 2025, we visited several companies in China and the west coast of the USA. What follows are some of the observations that stood out to us the most.

Beyond the headlines

Dr Yu’s comments on competition felt particularly resonant at the time of our visit to China. We had arrived nearly two weeks earlier in the midst of the delivery wars. The roads were filled with drivers, on mopeds and bicycles alike, racing around with one-yuan lattes and heavily discounted meals.

We had already seen this intensity begin to weigh on portfolio holding, Meituan’s margins. Yet when we spoke with management, they shared that while authorities had criticised this bout of irrational pricing, we should not expect decisive intervention in the near term. These battles are stimulating consumption and supporting near-term growth targets, and as a result, competition in consumer-facing sectors is being allowed to run its course.

It would be tempting, on the back of this experience and precedents in other industries, to infer that AI companies in China will face similar competitive dynamics, limiting the extent to which customer value translates into shareholder returns.

Yet, when it came to AI and semiconductors, the companies we met with emphasised that the government does not intend to repeat the experience of electric vehicles or other sectors where years of unconstrained competition destroyed significant capital.

At Biren Technology, a domestic GPU designer, management was explicit: “AI is not just a lifestyle technology; it is a matter of national security. The government cannot allow there to be missing pieces. It cannot fail.”

We also heard of concentrated efforts to encourage coordination and collaboration across the AI ecosystem, from foundries and equipment suppliers to chip designers and software layers. In practice, this means shared roadmaps and enforced interoperability, all in service of building scale and resilience in strategically critical technologies.

At the same time, competition remains a powerful force, and in some respects a constructive one. Across several companies, teams spoke about operating under constraint. Limited compute, scarce capital and constant competitive threat force prioritisation, speed and a relentless focus on efficiency.

Whether at Minimax, optimising model performance under tight resource budgets, or at Horizon Robotics, navigating supply-chain and foundry bottlenecks, this environment is shaping organisational design, product strategy, and competitive behaviour. This may ultimately distinguish China’s AI trajectory from that of prior technology waves, where abundance rather than perseverance set the tone.

The lesson from China was not to stop at the obvious (and first) conclusion. That same discipline proved just as valuable when we turned our attention to companies on the US west coast.

Where nuance beats a neat summary

When you look at a business posting nearly 70 per cent year-on-year revenue growth, you do not immediately assume there is an issue with its product development, and nor will ChatGPT. This is even less intuitive if revenue is a direct function of user frequency, which is up almost 20 per cent, as is the case with Reddit.

Yet our meeting with founder CEO, Steve Huffman, and chief operating officer (COO), Jen Wong, challenged this linear way of thinking. To our surprise, Steve opened the discussion by expressing acute dissatisfaction with Reddit’s product evolution.

His diagnosis was simple. As businesses scale, product managers become incentivised to introduce rules and processes designed primarily to avoid downside. Accountability for risk-taking diminishes, innovation slows and products begin to ossify.

They were clear that this is not unique to Reddit, citing several popular applications that failed to evolve at an appropriate cadence once they reached meaningful scale. Coincidentally, Adam Foroughi, CEO of AppLovin, shared this same observation with us earlier that week.

To counter this dynamic, Reddit has made material organisational changes. Several layers of middle management have been removed, and the product team now reports directly to Steve and Jen. The intent is to restore speed, energy and risk appetite.

This unusually candid conversation reinforced our investment thesis. Reddit’s resilience stems from its highly differentiated content, which occupies a valuable and defensible niche on the internet. At the same time, the platform has meaningful scope to improve how effectively that content is matched with relevant users - an opportunity that, if executed well, could materially enhance the long-term value of the business.

Over at Roblox, online gaming and creation platform, we similarly benefited from spending time with Dave Baszucki, the founder and CEO, and Naveen Chopra, chief financial officer (CFO), at their San Mateo headquarters. This meeting felt particularly timely given the shares are facing some short-term volatility.

Roblox had an exceptionally strong 2025 by almost any operational measure. Booking growth accelerated sharply and significantly ahead of guidance, driven by a wave of viral experiences. In fact, Dave noted that three games over the past two quarters each reached higher concurrent user counts than the entire Roblox platform did in the prior year. This appears to be as a result of improvements that were made to search and discovery, and in particular, changes to the recommendation engine.

At first glance, then, it might be tempting to assume Roblox is entering a simple phase of accelerating operating leverage. Yet our meeting highlighted why the near-term is likely to be messier. Several deliberate choices are being made that may create headwinds to reported growth and margins, even if they strengthen the platform’s long-run health.

The most immediate is safety. Roblox introduced broad age-gating from January 2026, making it effectively impossible to use the platform without verifying age. Management expects both a one-off engagement friction effect and the possibility of user-data revisions as misreported ages are corrected.

Infrastructure investment is another. As viral events scale, Roblox has been evolving its infrastructure to balance its owned data centres with third-party cloud “burst capacity” to support extreme concurrency without outages. That is strategically sensible, but it means dampening near-term leverage and some margin pressure to ensure the user experience holds as the platform pushes toward much larger peaks.

Understanding which headwinds reflect value-destructive issues, and which reflect intentional long-term investments in infrastructure, safety and the content flywheel, is essential if we are to be good holders through inevitable volatility. In other words, the meeting did not eliminate uncertainty, but it improved the quality of our judgement about what we are seeing, and why.

Reality checks

Across very different geographies and business models, a common thread reemerged from our recent trips. The most important insights were rarely the most obvious ones, and often many layers beneath headline metrics, consensus narratives or surface-level interpretations.

As artificial intelligence makes information more abundant, the risk of stopping at first conclusions only grows. Our experience this year reinforced a simple but enduring reality: perseverance, curiosity, and time spent with the builders of the future remain essential to separating signal from noise, and to identifying opportunities that may be mispriced by those unwilling to do the dirty work.

Risk factors and important information

The views expressed should not be considered as advice or a recommendation to buy, sell or hold a particular investment. They reflect opinion and should not be taken as statements of fact nor should any reliance be placed on them when making investment decisions.

This communication was produced and approved in February 2026 and has not been updated subsequently. It represents views held at the time of writing and may not reflect current thinking.

Potential for Profit and Loss

All investment strategies have the potential for profit and loss, your or your clients’ capital may be at risk. Past performance is not a guide to future returns.

This communication contains information on investments which does not constitute independent research. Accordingly, it is not subject to the protections afforded to independent research, but is classified as advertising under Art 68 of the Financial Services Act (‘FinSA’) and Baillie Gifford and its staff may have dealt in the investments concerned.

All information is sourced from Baillie Gifford & Co and is current unless otherwise stated.

The images used in this communication are for illustrative purposes only.

Important Information

Baillie Gifford & Co and Baillie Gifford & Co Limited are authorised and regulated by the Financial Conduct Authority (FCA). Baillie Gifford & Co Limited is an Authorised Corporate Director of OEICs.

Baillie Gifford Overseas Limited provides investment management and advisory services to non-UK Professional/Institutional clients only. Baillie Gifford Overseas Limited is wholly owned by Baillie Gifford & Co. Baillie Gifford & Co and Baillie Gifford Overseas Limited are authorised and regulated by the FCA in the UK.

Persons resident or domiciled outside the UK should consult with their professional advisers as to whether they require any governmental or other consents in order to enable them to invest, and with their tax advisers for advice relevant to their own particular circumstances.

Financial Intermediaries

This communication is suitable for use of financial intermediaries. Financial intermediaries are solely responsible for any further distribution and Baillie Gifford takes no responsibility for the reliance on this document by any other person who did not receive this document directly from Baillie Gifford.

10060715