A method for deciding on technology investment

The faster a technology evolves, the less we can reasonably expect experts to tell us with any certainty what will happen. If we look at artificial intelligence, cryptocurrencies, or cybersecurity (to name a few), we see people with unquestionable expertise reaching opposite conclusions. So decision makers, even if they don’t know who is right, must at least prevent uncertainty from turning into paralysis or igniting frantic race to chase after every new development.

There is no shortage of examples. Dario Amodei, CEO of Anthropic, is drawing attention to the possibility that frontier models could enable catastrophic cyberattacks, and arguing for slowing their development. Yann LeCun, a leading academic authority in the field, has expressed radically more skeptical views about the catastrophic risks attributed to AI. In the financial world, Jamie Dimon, Chairman and CEO of JPMorganChase, has repeatedly called Bitcoin a Ponzi scheme, while Larry Fink, Chairman and CEO of BlackRock, after being skeptical for years, has come to regard it as a legitimate financial instrument.

In any case, if even the experts are divided, we need to understand where a given technology stands today and where it is heading. This is the logic behind the HIT Radar, the tool developed by the DEVO Lab, now in its thirteenth edition, which assesses each technology according to three parameters: distance from adoption, impact, and speed of evolution toward commercial availability.

• Distance from adoption indicates how close a technology is to being used in practice by organizations. A technology may still be in the research stage, or under experimentation, or already available on the market. The distinction is essential because even if something is technically feasible, that doesn’t mean it’s ready to be adopted: Mature products, skills, infrastructure, standards, or simply sufficient demand may still be lacking.
• Impact captures the change a technology could bring about. A solution that is relatively close to going to market may have limited consequences, while a technology that is still a long way from adoption may promise to profoundly reshape processes, business models, or entire markets. Distance and impact, therefore, are not the same thing. And keeping them separate helps avoid a common mistake: confusing a major technological potential with managerial urgency.
• The third dimension, speed, makes the picture a dynamic one. It measures how rapidly a technology is progressing toward commercial availability. This is the parameter that helps determine when to start taking action: A technology that is still experimental may not require immediate investment, but it deserves attention if development is accelerating. Conversely, a technology with enormous potential impact may have years to go before it can be of any practical use.

It is the combination of these three dimensions that produces useful information for decision-makers. A technology that is close to adoption, high-impact, and evolving rapidly clearly requires the most attention. But even a technology in its infancy today can become strategically relevant if it starts maturing rapidly. The Radar approach helps identify which signals to watch for, and when a technological possibility begins to turn into a managerial issue.

Looking at emerging technologies in this way gives us results that are less straightforward, but more useful, than the diametrically opposed enthusiasts and skeptics. In the cryptocurrency world, for example, stablecoins already demonstrate a practical application of blockchain in payments, particularly cross-border payments. By contrast, still at an exploratory stage is the tokenization of real-world assets (the ability to digitally represent and trade stocks, bonds, or other assets): Technological development is happening fast, but sufficient market demand is still lacking.

Something similar is taking place in artificial intelligence. Local AI, meaning the set of technologies that allows models to be used and governed within an organization’s own infrastructure, can offer greater control over data, less dependence on cloud providers, and, in some cases, economic advantages too. This type of AI is still largely experimental and entails greater management complexity, but the pace of development is high. There is no need to decide today whether the future will be in the cloud or on-premises: What matters is recognizing if and when the balance between the two alternatives shifts.

Cybersecurity convincingly clarifies the importance of speed. The time it takes to discover a vulnerability and then exploit it has shrunk dramatically, in part because of AI. At the same time, defensive technologies with varying levels of maturity and impact are advancing: from confidential computing, which protects data while it is being processed, to architectures designed to manage the growing number of non-human identities, to post-quantum cryptography. The latter appears more distant, but it does require organizations to think ahead, because some actors are already stealing data in the hopes of being able to decrypt it in the future using quantum computers.

These examples show why a technology cannot simply be classified as important or irrelevant, mature or immature. What matters is the trajectory. One technology may have limited impact but move toward the market very quickly; another may have the capacity to generate enormous consequences but remain far from adoption; a third may already be available but make sense only for narrowly defined use cases.

In a context marked by strong disagreements even among experts, companies need the ability to organize uncertainty by building a system that allows them to see when the future is just around the corner: how far away adoption is, how great the impact could be and, above all, how quickly that distance from development to adoption is shrinking.