.svg)
Dear Investor,
One of our investment team members was visiting San Francisco last week and was reminded of the quote attributed to William Gibson, “The future is here - it’s just not evenly distributed.” During his few days in San Francisco, he nearly collided with robots rushing through his hotel delivering items, navigated intense security to visit a portfolio company protecting its unique IP, and had the choice of whether to get from place to place in an “old school” rideshare like Uber or Lyft or choose between competing robotaxi services offered by Waymo and Tesla.
Over the course of a week in early January, we saw announcements of new healthcare product launches by both OpenAI and Anthropic, the release of Cowork by Anthropic, the launch of Nvidia's next-generation Rubin platform, as well as many other product launches. This underscores the bewildering pace of change that we are living through.
The future is here, but it won’t be uneven for long
In one sense, that future won't be uneven for long; those technologies and much more will soon be prevalent in big cities and small villages around the world and utilised by everyone from the largest enterprises to individual consumers. However, in another sense, the future is more uneven than ever. The lack of symmetry between the relatively small number of people shaping the future and those impacted by the future raises complex questions. The impact of AI will be so profound that we are all having difficulty envisaging the future.
The changes heralded by AI innovation will, in many ways, be remarkably exciting: we will be much smarter and have the capacity to do things we couldn’t previously dream of; we will be much healthier, and unimaginable abundance will be part of everyday life. However, some of the impacts of AI will likely be profoundly negative; will we have a job and where we will obtain both our income and purpose for those who currently obtain much of their purpose from their work. These questions would create complexity in any environment, but will be particularly so in the highly polarised and charged world we live in today.
In the future, machines will almost certainly produce “better” music or art than humans, and robots will drive cars faster than Formula 1 drivers and beat the best golfer or tennis player in the world. However, we can be confident that people will prefer music or art created by humans and will want to watch a sporting event involving humans and not machines. We think people's desire to partake in these activities will be greater than ever in an AI-first world, and our desire for human connection will come to the fore. As an example, machines first beat humans at chess in the late 1990s, but this has done nothing to diminish the level of participation or interest in chess.
Despite our desire for human connection, incentives to optimise for price and/or quality will likely mean that we will choose to drive in a robotaxi, we will buy manufactured items created solely by robots rather than humans, we will obtain legal advice from a machine and yes we might decide that a machine does a better job of investing in startups than a human. What does this mean for the manufacturing worker, the taxi or rideshare driver, the lawyer or the venture capital investor?
We might debate how long these changes will take and whether AGI is genuinely just around the corner, several years away, or perhaps even longer. We might also debate what we mean by AGI. Truthfully, it doesn't really make a big difference whether some of these changes take one or two years, or five or six years or more. The core point is we have some fundamental questions to ask as a society as we enter a period of technological and societal change that is so profound and so fast, and for which there is no proper historical precedent.
The only certainty is change, and the prize to win is larger than ever
We spend a lot of our time discussing and debating these issues and recognise that the future will play out in very uncertain ways and that some of the scenarios described above may be completely off the mark. We never lose sight of the fact that our job is to invest in companies that will shape the future and create value for you as Limited Partners in our funds. We have been transparent that our jobs are much harder today than they have been in the past, but the prize is so much bigger. This new environment motivates and excites us.
As we look at our current portfolio, it includes a mix of companies operating in different layers of the stack. In some cases, these companies operate at the foundational model level (e.g. Thinking Machines Lab) and in other cases, they provide tools or enablement so that other organisations can drive this change (e.g. Supabase, Qodo or Deci, which was acquired by Nvidia).
However, most of our investing activity is at the application layer, including companies such as Aidoc, Sumble, Cuttable, and Lorikeet.
We believe industry by industry and market by market will be transformed by new AI native applications over the next few years, and we are in the very early days of that evolution. This focus on the application layer is very consistent in many ways with the sort of investing we have done over the past 13 years in SaaS businesses such as Canva, Neara and Vend. In other respects though, the emergence of these AI-native businesses throws up a whole series of new questions, which means we are both drawing on learnings from the past 13 years as well as doing a lot of first-principle thinking and analogising with other technological eras.
Over the past six months, the aggregate value of our portfolio has grown a remarkable 24%. This is the result of strong fundamental performance of many of our portfolio companies, and numerous companies undertaking capital raises at valuation uplifts. In the report for each fund, we have set out the fund's performance as well as the performance of the individual portfolio companies. We are thrilled with the performance of our portfolio, both in aggregate and for each of our funds individually. Most of that strong performance is driven by our largest companies, but we also have a clear line of sight to our next generation of companies which we believe will increasingly drive fund performance over the next 5+ years.
It is useful to reiterate our philosophy in creating value for you. To be clear, the approach is relatively simple, but the execution is much harder. As always, the devil is in the details.
- Each year, we invest in 7-12 new companies across our core themes and geographies. These companies will typically be pre-revenue and, in many cases, pre-product market fit. In each case, we are making the investment because we believe the combination of founder, theme and timing offers the possibility of a remarkable business.
- Each venture fund will be made up of 20-25 companies. The spread of portfolio companies in a fund serves as a risk mitigant, but we are very disciplined about investing only in those companies that we believe have the potential to be outlier companies. A portfolio of poor investments does not mitigate any risk.
- Approximately 50% of the capital in each fund is reserved for follow-on investments. We need to both exercise good judgment and be very disciplined about ensuring that these dollars are deployed to the best-performing companies. If we have 25 companies in a fund, then on average (which in reality never really applies), each portfolio company will initially have a 2% allocation in the fund. If we do our job well in allocating the remaining 50% of the fund, then the best few companies will each have an allocation of up to 10% or more of the fund.
- Once our venture funds are exhausted in terms of their capital, our Opportunities Funds continue to back the best-performing companies over an extended period. This enables us to maximise returns from our best portfolio companies and maintain long-term relationships with them.
- Across our portfolio, there will be a very small number of genuine outliers that not only have the capacity to deliver exceptional returns from the first investment we make, but also continue compounding at a fast rate over a lengthy period. When we find these companies, we want to do two things: deploy as much capital as we practically can in these companies and hold these positions for as long as practicable. Possibly the worst sin, and most expensive mistake, in venture capital is selling your best portfolio companies too early or failing to deploy enough capital in them.
- We are very conscious that you have a desire to have capital returned to you in a reasonable timeframe and at multiples of your initial investment. As well as the cohort of genuine outliers, there is a larger group of companies that have performed extremely well but may not compound at quite the same rate as the very top tier. These companies continue to be very attractive investments, but do not have the same asymmetrical risk to the upside. Our job is to proactively exit these positions at the right time and ensure we return capital to you, not just at the end of a fund's life, but in earlier years as well.
We hope the articulation of our approach is both helpful and resonates with you. We take our role as stewards of your capital very seriously, and we know that venture capital entails extended periods of illiquidity, which leads to an expectation of strong returns.
Team update
We want to thank James Tynan, who concluded his time with Square Peg at the end of 2025 after five years with the firm. Since joining in 2020 and becoming a Partner approximately two and a half years ago, James has been an important contributor at Square Peg. We wish him every success in his next chapter.
Our team is very well-resourced with an intentional mix of skills and experience. Over the past year, we have added four new hires across our investment team in Australia and Singapore. We are also currently looking to hire an additional team member in Tel Aviv.
