AI value creation ≠ AI value capture

Despite predictions of explosive growth in the AI market, current investment trends reveal that true economic transformation is lagging, with most capital focused on digital disruptions among tech giants rather than foundational changes in traditional industries. This approach risks redistributing digital profits without generating substantial new value for the broader economy, highlighting a critical need for operational expertise and long-term commitment to drive genuine productivity gains.

Fortune Business Insights™ projects the global AI market will grow to USD 1,771.62 billion by 2032. But don't let the figures fool you - AI won't transform the economy organically.

The more probable scenario is a high-stakes game of musical chairs among Big Tech: OpenAI cannibalising Google, Meta and Amazon... rather than real, meaningful transformation of operations and industries that will really benefit from it, like manufacturing, healthcare, logistics and financial services.

Investors and markets are overwhelmingly betting on digital disruption because it is faster and easier (when embedded successfully). But the implications for economic productivity, equity markets and passive investors are profound, and mostly ignored.

The overheating server in the room…

The generative AI investment thesis hides a blind spot. We are pouring billions into foundation models, while simultaneously ignoring the elephant in the room: very few companies are using this technology to transform the real economy. Instead, the smart money appears to be betting on something much simpler: OpenAI displacing other, larger, established tech businesses.

This reveals a paradox:

The theoretical disruption potential spans every industry - yet capital and leadership to own transformation in more traditional industries remains conspicuously absent.

The Path of Least Resistance

Investment behaviour suggests a collective belief that the route of least resistance lies in existing digital business models, rather than physical ones.

The logic: why invest in the complex, capital-intensive task of deploying AI across fragmented industrial sectors when you could simply back whichever technology displaces the digital monopolies dominating the S&P 500?

OpenAI doesn’t need to solve supply-chain optimisation or clinical diagnostics to generate large value. It just needs to be better at search than Google, more engaging than Meta or more useful than Amazon’s recommendations.

These are digital problems with digital solutions.

For more traditional industries, the economic incentives of frontier model developers are not necessarily aligned with the interests of the real economy.

Demolishing Google’s advertising revenues won’t help manufacturers become more productive or supply hospitals with better outcomes. At best it might temporarily reduce advertising costs, before a new digital monopoly emerges. We’re not talking about economic transformation; we’re talking about rent-extraction with a new landlord.

Nobody's Talking About Deployment

And this explains the curious absence of serious enterprise AI-deployment discussion. Everyone talks about foundation models and capabilities, but almost nobody talks about the unglamorous and complex work of implementing, embedding and maintaining these tools in operating companies.

The venture-capital money flowing into AI startups isn’t being directed toward integration partners, change-management specialists or sector-specific deployment vehicles. Instead it’s going to those who can accelerate disruption of digital incumbents.

From a pure returns perspective, this behaviour makes sense. Displacing Google’s search monopoly or Meta’s attention economy offers faster, cleaner exit opportunities than transforming how manufacturers optimise production lines. The total addressable market is massive, business models are proven and timeframes are shorter.

The Uncomfortable Truth

We don’t really know what the big tech are planning. If they move aggressively into advertising, search monetisation or other revenue models directly competing with FAANG business models, they could instantly undermine the valuations of companies which collectively represent trillions in market capitalisation.

The Magnificent Seven technology stocks aren’t just large holdings, they’re foundational to indices, pension funds, retirement accounts and the broader wealth effect underpinning consumer spending in the U.S.

If OpenAI successfully executes, we’re not talking about creative destruction that builds new value - we’re talking about substitution. Their displacement of current monopolies wouldn’t create net new economic activity; it would simply redistribute existing digital rents - potentially triggering a large market correction.

The Reality of Transformation

Perhaps this is why the 'build the AI and transformation will happen' narrative persists. It’s comfortable. It’s optimistic. It’s almost inevitable in this day and age.

But the reality is, the current investment patterns suggest we’re betting on digital musical chairs rather than genuine transformation. In the long run, AI will transform physical industries but that transformation demands patient capital, deep sector expertise and a willingness to wrestle with hard implementation problems.

At QuantSpark, we’re increasingly asked by clients across different industries to assess AI value-creation opportunities. We embed real transformation into businesses across the global. And this requires deployment capability, not just technology licensing. The returns are there, but they demand the kind of operational focus and committed capital that seems increasingly unfashionable in an era of foundation-model hype.

So here’s the question - do you want to bet on who wins the battle for digital rents or do you want to invest in actually transforming how your business, industry and the economy works?