Trying to Decide What to Do with AI? Think like a Venture Capitalist

By Geoffrey Moore

Author – The Infinite Staircase: What the Universe Tells Us About Life, Ethics, and Mortality

If you and your colleagues have embraced the AI mandate but are struggling with how best to get started, you have a lot of company.? High-risk, low-data decision-making is not something they teach in business school, but technology disruptions force the practice nonetheless.? The good news is there is one industry that does this sort of thing for a living—venture capital.? They have developed some best practices that help guide us all to better outcomes.

VCs think in terms of rounds of funding, with each round designated to take a particular kind of risk off the table.? A normal sequence is technology risk first, then market risk, then team risk, then financing risk—the last teeing up what one hopes to be an attractive exit.? Enterprise management teams can follow the same progression when engaging with a disruptive innovation like AI.? Here’s how it might play out.

Technology risk.? The good news here is that this is not your problem.? Until the vendors have taken this risk off the table, you can simply ignore them.? Of course, knowing when you can trust their claims of having done so can be a bit of a challenge, but I think it is fair to say with respect to AI that this stage gate has been passed.? It is the next one you need to be looking at now.

Market risk.? The kind of risk you want to look at now is what the VCs call product-market fit.? What is the use case that creates a return sufficient to warrant the risk of investment?? The point is, this is not about the technology, and you should not give this problem to a technologist.? You are looking instead for trapped value, most likely in your operating model, the sort of thing your industry has lived with forever but now, with the advent of AI, has the potential to break free from.? The most valuable use cases will likely show up in your Performance Zone , allowing you to break free from constraints that lessen your ability to serve your customers.? Absent one of these, there are any number of internal use cases in your Productivity Zone that will return a good ROI—just think of all the stupid stuff you put up with today.? Finally, while these internal use cases are easier to take on, failing to modernize your Performance Zone can leave you at a competitive disadvantage that will be tough to recover from, so even though the risk of commission is higher here, it is offset by an even higher risk of omission.

Team risk.? This is by far your biggest risk because the IT team you have in place was not recruited for this task.? It doesn’t mean they can’t take it on, but it does mean you have to vet the project leader with respect to their zone management leadership style.? Recall that the normal job of the Productivity Zone is to take time to reduce risk and reduce costs.? You do this by doing what Daniel Kahneman has taught us to call “think slow.”? That works fine as long as you are not facing a disruptive innovation.? When you are, however, you need to reverse the priority and take risks to reduce time, dismissing the cost because you are not yet at scale.? Now, you need to “think different.”? In particular, you want the team leader to bring a beginner’s mind to the target use case in order to find the optimal angle of attack, given the unprecedented capabilities AI is now demonstrating.

Financing risk.? The good news here is that unlike a VC, you don’t have to plough a lot of money into your project to get through a J-curve.? The challenge is that you do have to divert funds from your current operating model and that is at a time when it is increasingly being challenged by the next generation.? The Productivity Zone leaders, and this includes finance and HR as well as IT, will likely push back against such incursions, or at minimum, try to slow them down—because, as we said, their charter is to take time to reduce risk.? But in reducing the risk of commission, they are dangerously increasing the risk of omission.? So, if you have a promising risk case, your bet is to push back the other way and fast-track your effort.

That’s what I think.? What do you think?

Follow Geoff on LinkedIn | Geoffrey Moore Main Mailing List | Infinite Staircase Mailing List

________________________________________________________________

Geoffrey Moore | Infinite Staircase Site | Geoffrey Moore Twitter | Infinite Staircase Twitter | Facebook | YouTube