Crafting an enterprise AI strategy around generative or large language models is a bit of a high-wire act, mainly because of the lightning speed at which the technology evolves. Attempting to create a traditional transformation plan — the kind that culminates in an elaborate PowerPoint presentation meant to be executed over several years — is like trying to hit a moving target. The pace of tech advancements will always outstrip your ability to finalize the deck. So, what you need is a fresh approach to maneuvering this dynamic landscape. Luckily, we can borrow some principles from the US military.
Enter the OODA Loop, a strategic model devised by military strategist John Boyd. The acronym stands for Observe, Orient, Decide, and Act. This cycle is employed for quick and effective decision-making, especially in competitive scenarios. The ‘Observe’ phase involves data collection about the challenge at hand. During ‘Orient’, we interpret this information and understand its context. The ‘Decide’ step involves creating an action plan based on the collected insights. Lastly, in the ‘Act’ phase, we implement the decision. The cycle then restarts, using the feedback from our actions to observe alterations and fine-tune our strategy accordingly. This iterative process encourages adaptability and prompt decision-making in uncertain situations — a perfect fit for the world of AI strategy.
But how exactly can we apply the OODA Loop’s principles? Let’s roll back to the first half of 2023. Late last year, you battled through your company’s yearly (and probably painful) budgeting process. You drew up a plan, created roadmaps, and got started on executing them in Q1. Then, out of nowhere, GPT-4 came along on March 14, 2023, flipping the world on its head. Your annual budgeting process crumbled as you realized the need for immediate reprioritization. Currently, most companies are in a scramble, attempting to fund their GPT initiatives with off-cycle budgeting or project cuts.
Now picture a budgeting process underpinned by the OODA Loop.
Observe: GPT-4 is announced and within a month, a flurry of videos, papers, and articles discussing its potential are all over the place. Your organization gets to work learning about it and figuring out how to adapt to this monumental shift.
Orient: Now you need to contextualize this shift within your organization. Initial hypotheses about how it may open up opportunities and introduce competitive threats need to be drawn up.
Decide: This is where most organizations stumble — the actual decision making. Sure, you won’t have perfect information, but you can make a start. Identify potential use cases with key stakeholders, sketch out potential proofs of concept. The idea here is to make swift, incremental progress, even if the information isn’t flawless, rather than wait for the perfect moment.
Act: Time to implement. Execution is where big organizations often fumble, but repeated application of the OODA Loop principles will refine your system, frameworks, and teams for swift action and response.
Is this easy? Of course not. But what is your other option? Adapt or become obsolete. Technology isn’t going to slow down or stabilize. And there’s no way you can keep up if your strategy for 21st-century technologies is rooted in 19th-century processes. It’s time to adapt, evolve, and thrive.
Prolego is an elite consulting team of AI engineers, strategists, and creative professionals guiding the world’s largest companies through the AI transformation. Founded in 2017 by technology veterans Kevin Dewalt and Russ Rands, Prolego has helped dozens of Fortune 1000 companies develop AI strategies, transform their workforce, and build state-of-the-art AI solutions.