A defining feature of the AI boom has not just been the pace of technological progress, but the scale and speed of capital being committed in anticipation of future economic returns. As AI moves from experimentation toward large-scale deployment, the focus is increasingly shifting from capability to capital, with financing structures and return durability becoming central to investment outcomes. Against a backdrop of heightened volatility driven by trade policy uncertainty, global equity markets nevertheless delivered a third consecutive year of double-digit gains in 2025, with AI-related investments central to the rally. That performance, however, has increasingly been shaped by expectations of future returns rather than proven economics. As capital intensity and financing complexity rise, the AI opportunity set is becoming more uneven, elevating the importance of balance-sheet strength and disciplined stock selection.
Financing the AI Build Out
These underlying economic tensions became more visible in the second half of 2025, as a wave of high-profile partnership agreements revealed a shift in how AI growth is being financed. While scaling laws continue to hold, sustaining frontier AI performance now demands capital on an unprecedented scale. Monetization is advancing quickly yet remains insufficient to fund the level of investment required to stay competitive. The result is a widening disconnect between commercial revenue and capital requirements, pushing companies towards large-scale and increasingly debt-funded strategic partnerships.
These pressures have produced a clear divergence within the AI landscape. Hyperscalers such as Microsoft, Alphabet, Meta, and Amazon are funding the build-out largely from free cash flow, enabling them to scale models, infrastructure and distribution without significant balance-sheet strain. By contrast, many independent AI developers, including OpenAI and Anthropic, have entered into agreements of extraordinary scale and uncertainty to secure the infrastructure required to remain competitive. For example, OpenAI, currently loss-making, has entered into a reported $300bn multi-year contract to secure data-center capacity from Oracle, despite limited certainty that the agreement will be fully utilised.
Market optimists argue that comparisons with the dot-com cycle are misplaced, noting that that period produced “dark fiber”, infrastructure that went largely unused, whereas today all AI-infrastructure is being fully utilized. While utilization rates may be higher, this view overlooks a key economic risk: a meaningful share of GPU demand is driven by customers whose business models are not yet self-funding, with spend effectively underwritten by their valuations. It is also unclear how much demand reflects economically valuable workloads versus AI “slop” generation where end-user consumption is effectively being subsidized by companies to drive adoption and seed use cases. Should this funding loop weaken, these subsidized activities could evaporate, raising the risk of oversupply seen in previous bubbles.
Valuations and the Distribution of Returns
These economic uncertainties are increasingly visible in AI-related valuations. While the transformative potential of the technology is widely acknowledged, there remains limited clarity around the returns that will ultimately be generated on the large amounts of capital being deployed.
As a result, some of the more crowded AI exposures now appear increasingly exposed to valuation risk. Companies such as Nvidia, AMD, and a range of specialist data-center providers, alongside AI-enabled software names such as Palantir are increasingly priced for near-perfect execution, leaving limited margin for error should demand growth moderate or monetization not materialize.
By contrast, companies with strong balance sheets, established cash flows and diversified earnings streams are better positioned to invest through the cycle without relying on debt. Alphabet and Microsoft for example are funding substantial AI-related capital expenditure from existing free cash flow, minimizing the risk and giving them time to monetize AI across their broad platforms.
Outside the hyperscaler universe, more defensively structured paths to AI participation place greater emphasis on capital discipline. Capgemini’s exposure is focused on enterprise implementation rather than frontier model development, while Samsung combines balance-sheet strength with strategic positioning across semiconductors, memory and hardware. Both provide exposure to AI’s diffusion while avoiding the most capital-intensive segments of the value chain.
Conclusion
AI has already reshaped equity market leadership and returns, but the forces that will drive the next phase of growth are less certain. Near-term outcomes are likely to be influenced by a combination of capital availability, financing conditions and shifts in investor sentiment, rather than by technological progress alone. Over the longer term, however, fundamentals are likely to reassert themselves, with returns increasingly shaped by cash generation, balance-sheet strength and the ability to translate AI adoption into sustainable economic value.
In this context, Mondrian’s value-oriented fundamental approach which focuses on risk-adjusted returns, enables us to select attractively valued securities, including those with exposure to the buildout and adoption of AI when valuations and the range of outcomes are favorable.
Disclosures
¹ MSCI World Index total return, USD terms
Views expressed were current as of the date indicated, are subject to change, and may not reflect current views. All information is subject to change without notice. Views should not be considered a recommendation to buy, hold or sell any investment and should not be relied on as research or advice.
This document may include forward-looking statements. All statements other than statements of historical facts are forward-looking statements (including words such as “believe,” “estimate,” “anticipate,” “may,” “will,” “should,” “expect”). Although we believe that the expectations reflected in such forward-looking statements are reasonable, we can give no assurance that such expectations will prove to be correct. Various factors could cause actual results to differ materially from those reflected in such forward-looking statements.
Neither MSCI nor any other party involved in or related to compiling, computing or creating the MSCI data makes any express or implied warranties or representations with respect to such data (or the results to be obtained by the use thereof), and all such parties hereby expressly disclaim all warranties of originality, accuracy, completeness, merchantability or fitness for a particular purpose with respect to any of such data. Without limiting any of the foregoing, in no event shall MSCI, any of its affiliates or any third party involved in or related to compiling, computing, or creating the data have any liability for any direct, indirect, special, punitive, consequential or any other damages (including lost profits) even if notified of the possibility of such damages. No further distribution or dissemination of the MSCI data is permitted without MSCI’s express written consent.
This material is for informational purposes only and is not an offer or solicitation with respect to any securities. Any offer of securities can only be made by written offering materials. The information set forth herein is a summary only and does not set forth all of the risks associated with the investment strategy described herein.
The information was obtained from sources we believe to be reliable, but its accuracy is not guaranteed, and it may be incomplete or condensed.
It should not be assumed that investments made in the future will be profitable or will equal the performance of any security referenced in this document. Examples of securities will represent only a small part of the overall portfolio and are used to illustrate our investment approach. Any holdings are subject to change and may not feature in any future portfolio. More information on holdings is available on request.
Unless otherwise stated, all returns are in USD.
All references to index returns assume the reinvestment of dividends after the deduction of withholding tax and approximate the minimum possible re-investment, unless the index is specifically described as a “Gross” index
Past performance is not a guarantee of future results. An investment involves the risk of loss. The investment return and value of investments will fluctuate.
Mondrian Investment Partners Limited is authorised and regulated by the Financial Conduct Authority (Firm Reference Number: 149507). Mondrian Investment Partners Limited is also registered as an Investment Adviser with the Securities and Exchange Commission (registration does not imply any level of skills or training).
