Before building an AI strategy, most companies skip the one thing that makes it work: an honest, painful reckoning with what the company actually is. They reach for a tool, a trendy hire, or an off-site full of action points, and they wonder later why nothing moved. The reason is simple. You cannot build a real AI strategy on top of a story the company tells about itself. You have to strip the story away first.
The expensive off-site that produces nothing
Gather a room of people who have never seriously used AI, and you will get a very expensive off-site and still no strategy. You will leave with a deck of bullet points, each assigned to someone else, and nothing will change. You do not need an off-site for that. What you need is the opposite kind of room: the people who understand your company intimately, sitting with people who genuinely know how to build with AI in real time, querying the data and the models back and forth, and refusing to look away from what they find.
Three days of shedding before any strategy
The honest version hurts, on purpose. Day one, you dig into what the company really is, what it has actually done, where it failed, what it overlooked. Day two, you dig harder, and it gets more uncomfortable, the kind of uncomfortable where people want to go home. Day three, you stop being able to hide from the conclusions, because the analysis has no feelings and no political agenda. It is concrete, specific, and it will sting. You can ask AI to soften its tone, but if you need it softened, you are usually closer to losing the thing than you realise.
By the end of those three days you are not strategising. You are stripped of the comfortable fictions of corporate life. And that is the point. The reason your AI project is returning nothing is that it was built on those fictions in the first place. This is the same pattern as AI being a culture project before a technology project: the foundation has to be honest before anything built on it can hold.
The marriage that looks fine and is rotten underneath
It is like a marriage you cannot repair until you have the painful conversation where everything unsaid finally gets said. Skip that conversation, keep buying gifts and booking trips while you quietly resent each other, and the whole thing is rotten at the core. Most struggling companies are exactly that: a relationship maintained on appearances, with the real conversation never had. That is why the projects fail.
Shed the skin, then build
A snake sheds its skin so it can grow into what it is meant to become. Your company has to be willing to do the same, to call the old story what it is and let it go. Shed the skin first. Then build the AI strategy. Only then does the P&L actually start to move. If you want someone to lead a room through that reckoning without letting it stay comfortable, that is the work I do with leadership teams.
Frequently asked questions
What has to come before building an AI strategy? An honest, painful reckoning with what the company actually is. Most companies skip it, reaching for a tool, a trendy hire or an off-site full of action points, then wonder why nothing moved. You cannot build a real AI strategy on top of the story a company tells about itself; the story has to be stripped away first.
Why does the usual AI strategy off-site produce nothing? Because it gathers people who have never seriously used AI and ends in a deck of bullet points assigned to others, with nothing changing. What is needed instead is the people who understand the company intimately sitting with people who genuinely know how to build with AI, querying the data and models in real time and refusing to look away from what they find.
What actually happens in the three days of “shedding”? Day one digs into what the company really is, what it has done, where it failed, what it overlooked. Day two digs harder and gets more uncomfortable. Day three, the conclusions can no longer be hidden from, because the analysis has no feelings and no political agenda. It is concrete and specific, and it will sting.
Why compare a struggling company to a failing marriage? Because both are relationships maintained on appearances with the real conversation never had. Keep buying gifts and booking trips while quietly resenting each other, and the thing is rotten at the core. A company that never has the honest conversation is in the same state, which is why its AI projects fail.
What does shedding the skin mean for a business? A snake sheds its skin so it can grow into what it is meant to become, and a company has to be willing to do the same, to call the old story what it is and let it go. Only after that honest reckoning is the AI strategy built on solid ground, and only then does the P&L actually start to move.
Thomas Anglero is a Strategic AI Advisor, keynote speaker and author of Intro to Artificial Intelligence. He has delivered over 450 keynotes across 30 countries for organisations including IBM, the WHO, the World Government Summit and the European Commission. He founded the IBM Watson AI Lab for Cancer at the Oslo Cancer Cluster and closed over $500 million in enterprise transformation deals as CTO and Chief Innovation Officer at Cognizant.