Stop Betting on a Single AI Use-Case – Manage a Diversified Investment Portfolio Instead

Just over three years have passed since ChatGPT made its public debut, and whether or not you feel like generative AI turned investors on their heads yesterday or a lifetime ago, the fact remains that we’re still in the “burst” phase of this latest technological revolution. 

Consequently, this also means we’re entering the “stall zone,” which is historically marked by increased skepticism and recession fears (hence all the recent speculation about an AI Bubble).  But, just as in previous industrial revolutions, the economic models have changed, the tech is catching on, and it’s the early adopters that are going to come out ahead when the “boom” or massive rollout inevitably comes around. 

Unfortunately, most enterprises are currently setting themselves up for future irrelevance due to their misguided investments in “random acts of AI” rather than a true commitment to transformation. In order to successfully make it through the stall zone, businesses should treat AI like a diversified investment portfolio. 

The Quick Win Quagmire 

The reason most AI pilots fail to materialize into any meaningful change is largely because the implementations are haphazardly rolled out for the sake of doing something in order to feel like you’re keeping up with the latest tech. While they can (and do) result in a measure of positive ROI, which feels promising, they don’t actually move the needle in the long run. 

Why? Because they mistake motion for progress. The majority aren’t actually integrated into the core business and therefore lack meaningful KPIs or goals. Additionally, because they often rely on packaged solutions, they aren’t fully customizable, leading to inevitable plateaus in effectiveness. Finally, more often than not, when calculating the inference bill at scale, the entire business case collapses as costs skyrocket by ten or even a hundred times. 

Speed without alignment is just noise, but that doesn’t change the fact that the pressure to show fast results is absolutely real. Boards want meaningful AI returns immediately – not in three years – and CIOs and CTOs are caught in the middle. So we can’t pretend that speed doesn’t matter. Early wins are crucial. They build confidence, create momentum, unlock budget, and they prove that AI can create value without massive risk.

But quick wins have to be fast, low-risk, and anchored to real business outcomes to create momentum, which is what makes the diversified portfolio approach effective. 

Three Types of AI Investments 

The way I think about it, there are three categories of AI bets: quick wins, staged bets, and a few high-conviction, game-changing plays: 

1. Quick wins are low-risk, fast time-to-value, strong momentum builders. Think of implementations like automated customer support chatbots: relatively easy to roll out and lead to quick ROI. These bets often have near-term payoff, but tend to lack strategy. It is important to think about multiple quick win bets and focus on building overall momentum.
2. Staged bets are core initiatives that balance risk and meaningful business outcomes. Take, for example, ultra-personalized customer notifications and messaging with contextual triggers. Imagine a customer support agent that follows up with a client after a purchase to check-on how things fit, triggered by that client’s history of returns. Staged bets are medium-term investments that are often easy to pilot but hard to commercialize at scale. They should be approached with an ‘invest and learn’ mindset. Done right, commercial implementations of staged bets often drive significant value.
3. Game-changing bets are fewer in number and higher in risk, but are also capable of redefining how a company competes. For instance, an omni-channel chat-based assistant that’s aware of consumer data across the entire organization – perhaps even data available publicly. So, rather than just triggered notifications, customers have well-informed, dedicated shopping stylists (agents) present with each purchase. These interactions are consistent and multimodal – ranging from text to voice to video to virtual try-ons. You can think of game-changing bets as the 10x bets that have the potential to change the DNA of your company. They are high-risk, medium-to-long term pay-off, and usually include a high amount of planning and ramped investment.

In the same way that a diversified investment portfolio is designed to balance growth and stability, this portfolio approach allows leaders to generate short-term results while building momentum for longer-term, transformational change. Rather than getting stuck solely with quick wins that deliver minimal lasting impact, this approach leverages quick wins to fuel bigger, more meaningful initiatives.  

Having these developments happen side-by-side creates resilience and adaptability as the technology and business environment evolve, laying the foundation for the larger transformation of real game changers.

The economics of game-changing bets have shifted

One thing to note that makes the current technology trend different is that AI is quietly redefining what qualifies as a “game-changing” investment. 

Historically, these bets demanded significant capital, long delivery timelines, and high organizational risk before value could be realized. Rapid AI enablement has inverted that equation. The cost to build, test, and iterate has fallen dramatically, while speed to signal has accelerated just as fast. Enterprises can now validate transformational ideas in weeks instead of years, allowing leaders to de-risk ambition through learning rather than delay. 

This shift doesn’t reduce the importance of strategic conviction and organizational change, but amplifies it. When speed and cost fall, the advantage moves to organizations that can learn fastest, not those that plan longest.

Commit to turn chaos into compounding value

Though diversified investment portfolios can work for just about everyone, the actual investments are going to vary greatly depending on the available resources, values, and financial goals of the owner. Which also happen to be necessary considerations for executive teams making AI investments, and frankly, might mean that some teams come to the more honest conclusion that they’d prefer to forego AI investment altogether. This is, after all, another path towards cost savings that could help avoid a whole lot of unnecessary effort.  

When discussing what it takes for an enterprise to truly unlock the massive opportunities that AI is capable of (i.e. true digital transformation), I think of a story from a much more analog time. In 1519, after taking over the city of Veracruz in present day Mexico, Spanish Conquistador Captain Hernan Cortes caught wind that some of his men were planning to abandon their mission and flee on a stolen ship. To make this impossible, he ordered that the ships be destroyed, forcing them to commit to the mission. 

Burn the ships. While it may sound a bit extreme, this is the level of commitment that meaningful disruption requires – there is no option to turn around and go back to the way things were. And this is not possible without a truly aligned leadership team that recognizes that AI is not magic, but an incredibly powerful tool that affects people, culture, and jobs, and thus requires clear business objectives, a plan for change management, and a willingness to rethink how work gets done.

So business leaders that are serious about maximizing the value of their AI investments have to start leading the change by first defining and agreeing on the long term vision, their “North Star,” and what prioritized game-changing bets will get them there. From there, every AI initiative must be anchored to a competitive advantage with measurable value. And finally, diversify the resulting AI investments across quick wins that build momentum while balancing staged and big bets.