I’ve sat in enough AI pitch meetings to know what the left-hand picture looks like in the deck. Model on top, data on the bottom, an arrow between them, a target date in the bottom right.

The project starts, and the team finds the right-hand picture.

Each of those layers is real work that has been deferred for years, because there was never a reason to fund it that the board cared about. Now there is one, complete with a deadline and a sponsor who built the timeline by counting to two.

And there’s Tom at the bottom, holding it up because half of his work was never officially anyone’s job and he’s the one who knew where the bodies were buried. The day Tom takes proper holiday, every layer above him becomes visible at once.

The model is the easy part. Pretending the other thirteen layers are not really thirteen layers is how AI programs end up three years late, and how heads of data end up handing in their notice.

I wrote a free playbook on fixing governance and ownership in 30 days: https://lnkd.in/dRkqe64E

♻️ Repost if your AI deadline was set by someone counting two layers.