Artificial intelligence has become the defining investment theme of this cycle. Capital is flowing into the sector at scale, and a small group of listed technology companies is absorbing a disproportionate share of that inflow.

In its latest market overview, Hamilton Lane, a U.S.-based private markets investment firm (with a Swiss sales office) overseeing about $1,000 billion in assets for institutional clients, describes AI as «the only investment question you need to answer». At the same time, the report raises a more practical issue: how to translate that exposure into returns in a market where access itself is unevenly distributed.

The «OpenAI trap»

While the report does not use the term, its argument points to a dynamic that could be described as an «OpenAI trap»: a large amount of capital converging on a limited number of highly visible AI companies, often clustered among a handful of dominant technology players, many of which remain private or are already valued at scale.

For public market investors, this often translates into indirect exposure through a handful of large-cap stocks. For private investors, it can mean competing for allocations in oversubscribed funding rounds at elevated valuations.

In both cases, the challenge is similar: access is constrained, and the price of entry reflects that.

Concentration in public markets

According to Hamilton Lane, AI exposure in public markets is highly concentrated. Index-based investors, in particular, are increasingly reliant on a small number of companies — and technologies — to capture what is often presented as the AI theme.

This concentration has supported strong performance. At the same time, the report suggests that it limits diversification and may reduce the scope for differentiated returns as more capital targets the same assets.

Where value is created

Hamilton Lane points to private markets as a broader access point. Venture capital and growth equity strategies invest across a wider range of AI-related activities, including applications and supporting technologies.

Crucially, the report notes that companies are staying private for longer. As a result, a significant portion of value creation may occur before a public listing, leaving later investors with a different risk-return profile.

This does not necessarily imply higher returns in private markets, but it does suggest that timing of entry has become more important.

A second constraint: liquidity

At the same time, the report highlights a separate challenge. Across private markets, distributions have remained weak, with investors waiting longer to receive capital back. At a time when private credit is under growing scrutiny, the report indicates that the segment has shown more consistent distribution patterns than other areas of private markets.

This creates a tension. The same market segments that may offer broader access to AI-related opportunities also require longer holding periods and provide less immediate liquidity.

Hamilton Lane presents this as a structural feature of the current environment rather than a temporary dislocation.

An evolving investment question

The report’s conclusion is less about endorsing a single approach than about structuring the investment question more coherently. Artificial intelligence may be the central theme, but extracting returns from it depends on where investors can gain access — and at what stage.

Given the concentration of AI exposure in public markets and the shift of value creation into earlier, private stages, the report suggests that many investors may be participating in the theme only after a substantial part of the upside has already been realized.

It points to private markets as one way to access earlier stages of that value creation, while also noting the longer holding periods and reduced liquidity involved.