AI transformation is a budget line, not a chatbot

AI transformation is the phrase boards use when they do not want to say "we will change how people spend Tuesday." I do not blame them. Transformation sounds expensive and official. Tuesday sounds like work.

An AI transformation strategy that cannot name the first workflow is a press release. An AI transformation roadmap with four colours and no owner is a poster. I have watched both survive a full planning cycle. Nothing moved except the vendor.

The useful version is boring. You pick a process that already hurts. You fund AI training for the people who live inside it. You give someone the right to kill the old way. You measure hours, errors and cycle time. Then you pick the next process. That is the roadmap. Everything else is scenery.

Strategy without the maturity model

Leaders love maturity models because they delay a decision. You score yourself a 2. You aspire to a 4. You hire a consultant to explain the gap. Meanwhile the claims team still copies data between two systems by hand.

Write the strategy as a sequence of bets. Bet one: this workflow, this owner, this kill test, this date. Bet two is not allowed to start until bet one has a residue. If that feels too small for a board pack, the board pack is the problem.

A Chief AI Officer can help if they have money. A CAIO with a title and a listening tour will produce a town hall. The appointments we cover are the ones that fund literacy in the first 120 days. That is AI transformation you can audit.

Vendors will offer you a platform that "enables" transformation. Platforms are fine after the work is named. Before that they are a second job. I would rather see a two-day AI design sprint on a live problem than a six-month implementation that nobody asked for.

Where it is actually happening

Not in keynotes. In hiring clusters, tenders and programme stand-ups. A HK or SG bank staffing AI-for-teams is a local signal. A UK agency buying design thinking is a different signal that still tells you the organisation is trying to change how it frames work. Digital transformation Australia, Singapore, Hong Kong: the language in annual reports is getting louder. The residue is still uneven. We write the residue.

Do not inflate a hiring spree into a national story. Do not ignore it either. The people who run innovation in those markets already know which peers moved. Naming the company is how you make the piece useful.

Digital transformation is the wider problem. AI is one of the tools. If your digital programme has no AI bet, you are still running 2016. If your AI programme has no digital seam (the customer, the process, the system you will switch off) you are running a lab.

Collective Campus still trains the rooms that have to live with this. The AI skills programmes and the wider innovation training are how a strategy becomes a habit. Two links. That is the offer. The rest of this page is the argument.

What we publish here

When a CAIO funds AI training is the appointment story. When a HK or SG bank staffs AI training is the market signal. AI training without the slideware is what the work looks like when you skip the costume.

The AI training playbook is the field guide. Infinitev's landing-page generator is a case of using AI to test ideas fast, not to decorate a transformation narrative.

If you are writing the roadmap this quarter, start with one hated workflow and a date. If you cannot name those two things, you do not have a transformation. You have a theme.

Boards will ask for a competitive scan. Give them three named peers and what those peers actually staffed, not a quadrant. A HK bank hiring AI trainers is more useful than a global heat map. A CAIO who funded literacy in 120 days is more useful than a prediction about AGI. Competitive scans that stay in the clouds are how transformation offices stay busy. Name the account. Name the move. Then decide whether you are copying the hire, the training or neither.