DEER ISLE: Insights, Flows, & Investment Trends
WILL AI DISPLACE SAAS? Not So Fast – Scale is Not Free!
The recent decline in SAAS stock prices and fears about SAAS private capital lending is generally acknowledged to have been triggered by the Citrini Research scenario — in which agentic AI triggers a human intelligence displacement spiral, collapses the ARR assumptions underlying trillions in private credit, and ultimately cracks the $13 trillion mortgage market. Its internal logic is sound, and its risk framing is useful to understanding “AI’s broad-based impact on the economy.
But it rests on the premise that the cost of deploying capable AI will fall fast enough that any sufficiently motivated enterprise can simply replicate mission-critical SaaS functionality in-house, eliminating the need for the platforms that billions in private capital are underwriting.
The counterargument begins with a simple economic observation — scale is not free, and intelligence at scale is extraordinarily expensive.
A Fortune 500 company deploying a self-built, enterprise-grade alternative across 50,000 users — with the reliability, compliance, audit trail, integration depth, and continuous improvement of a mature SaaS platform — is a fundamentally different undertaking than an individual or small team deploying a vibe coded solution.
The deeper flaw in the displacement thesis is what it assumes about the nature of AI capability itself. The Citrini scenario treats AI as a commodity that becomes universally accessible as inference costs fall — a world where the marginal cost of intelligence approaches zero and therefore no one needs to pay a platform to provide it.
What about real costs such as the cost of power? AI is currently being run on subsidized utility power as well as by private capital investors who are supporting data center and AI buildout in companies or projects which are not profitable. The IEA projects U.S. data center electricity consumption will grow 133% to 426 TWh by 2030, and Goldman Sachs estimates $720 billion in grid spending will be needed through that same period.
To use a minimal amount of AI computing power to answer simple questions like what to have for dinner tonight is free but to ask it to analyze a large amount of data has a high transaction cost. Each time the data is analyzed, the cost reoccurs (unlike asking about what to have for dinner tonight – ask away – it’s all free) which means that experience is valuable since it reduces testing costs.
Unique, proprietary insight is the real value driver of AI. It drives competitive advantage, informs institutional investment decisions, optimizes complex supply chains, or generates genuinely novel scientific output and requires massive scale: large proprietary datasets, fine-tuned models trained on domain-specific corpora, inference infrastructure capable of running at enterprise speed and reliability, and the organizational capability to translate model output into workflow.
These are not insights that a competent developer solves in weeks with an agentic coding tool. They are multi-year, capital and resource intensive infrastructure investments that only a small number of companies can make effectively.
This reframes the valuation question for SaaS entirely — and it is here that the McKinsey Rule of 40 research becomes the critical analytical lens. McKinsey’s study of over 200 software companies found that businesses exceeded Rule of 40 performance where revenue growth rate plus free cash flow margin equals 40 or above only 16% of the time, and that barely one-third of software companies ever achieve it at all. Yet the reward for those that do is dramatic: top-quartile Rule of 40 companies earn nearly three times the valuation multiples of bottom-quartile peers.
If SAAS companies are able to use AI to more efficiently achieve the Rule of 40 by increasing efficiency while increasing growth, their valuations should dramatically increase.
The businesses that are neither critical infrastructure nor Rule of 40 category leaders, underwritten on growth assumptions that were generous before AI and are now simply wrong, are precisely the daisy chain of correlated bets that Citrini’s scenario describes.
—————————-
Are you buying email lists of potential capital counterparties for deals or transactions? Do you really want to be spending you or your team’s time curating and figuring out which data is valuable? You can, within 2 weeks of signing up with us, be contacting your relevant capital from a universe that represents 80%+ of US fiduciary, institutional capital and receive important analytics to create a virtuous cycle from your outcomes. Email us at info@deerislegroup.com to learn more
—————————-
Capital Provider Interest: There is growing interest among capital providers in healthcare and biotech companies that can demonstrate traction with family offices and venture capital investors.
US Government Mandates and Focus: There is a strong push to attract U.S. government funding for projects related to critical minerals and strategic infrastructure. Reflecting bipartisan support for large-scale manufacturing investments that underpin national and economic security, the Office of Strategic Infrastructure’s financing mandate has expanded from approximately $1 billion to $200 billion. This now spans 31 critical technology categories and supply chain initiatives. For example, we have heard of a Philippine copper mine attracting investment, highlighting that capital allocations are increasingly focused on securing supply chains wherever they are located globally.
Venture Capital: A major question for the venture capital industry is how the decline in public SaaS stock valuations will affect SaaS companies within VC portfolios. According to 2026 data from Tracxn, there are approximately 23,800 funded SaaS companies in the United States. Collectively, these companies have raised about $923 billion in venture capital and private equity funding, with more than 14,700 of them having secured Series A funding or later.
Real Estate: The current market environment is challenging. As an indication of this, we are seeing real estate transactions offering preferred equity returns of 12–15%, particularly for deals that are not considered “down the middle of the fairway.”
