According to the Gartner Hype Curve, a descent into the Trough of Disillusionment follows the Peak of Expectations, but I’m not sure generative AI got the memo. It produced unprecedented expectations, to the point that many have predicted it will achieve human-level general intelligence that could even mean the end of civilization as we know it. Those expectations have been scaled back, at least by many, and we’re certainly now… somewhere… but I certainly wouldn’t call it a Trough of Disillusionment. Let’s call it a Slough of Confusion. What to do? MIT produced a study saying 95% of AI efforts don’t get past the pilot stage… but Jack Dorsey just announced that AI meant he could cut the work force at Block, his financial technology company, by 4,000 employees, or half the total. Lots of senior managers say they see productivity gains from gen AI… but lots of lower-level employees say the gains are illusory because they’re having to spend so much time supervising the AI and fixing the problems it causes. Businesses talk about harvesting low-hanging fruit… but Gallagher just released a study saying businesses are realizing it will take them two to three years to get the full benefits of the AI efforts they’re pursuing. When things would get hairy as a deadline approached and the shouting started, an old boss of mine would often walk through the newsroom, smile and call out, “Good luck in your chosen profession.” That’s sort of how I feel now: Good luck to all of us as we sort through the confusion on AI. But there are clearly things we need to be doing to eventually achieve clarity, two of which are key points that Dr. Michael Bewley of Nearmap hits in this month’s interview. One is hard but simple: Get going. Now. Even though it’s not clear just where to start or where you’ll end up, you’ll never get to the destination if you don’t start—and your competitors are surely underway. As Bewley puts it: “Gen AI opened up a new world. It is absolutely revolutionary. I think it's on the level of the internet being invented or the personal computer. So you definitely don't want to sit by and say, ‘Well, I'll wait and see what happens,’ or ‘This one's not for me.’ You've got to get involved.” The second is to go after that low-hanging fruit, even if Gallagher is right that it may take some time to get the full benefits. In Nearmap’s case, that means enhancing its existing capabilities by using AI to process aerial imagery more accurately and as quickly as possible—speed being of huge importance to both insurers and the insured as natural catastrophes unfold. We’ll still be in the Slough of Confusion for some time, I’d say, but we can at least start finding the paths that will take us out. Cheers, Paul | 
