They just made you Chief AI Officer. Or Vice President of AI. Whatever title they invented when the board asked about AI strategy.
Here is what nobody told you. This job will be great for your career and probably end badly. Not because you will fail, but because they do not know what they hired you to do.
That is fine. You are building the resume that gets you hired somewhere better.
You need to fix your compensation first. You make one hundred seventy thousand dollars now. They want to promote you. Ask for two hundred forty thousand dollars base plus a fifty thousand dollar signing bonus.
Why they will say yes. If they do not promote you, they will pay a search firm fifty thousand dollars, waste six months interviewing, and hire someone who does not know the organization. That person will fail in eighteen months anyway.
Your signing bonus costs less than the search firm. It does not create drama with other vice presidents. The math works.
Also demand fifty percent equity acceleration on change of control. Demand one hundred percent acceleration if they reorganize your role in twenty-four months. Require three specific success metrics tied to equity, including agent adoption, cycle time reduction, and cost savings. Finally, get twelve months of severance if they kill your role or cut your authority.
If you are still in this role in two years, you missed the window. Successful AI leaders move to companies that already get it and pay multiples more.
Think about what you are actually building. You are not transforming the company. You are building the case study for your next job.
Every decision should answer how this will sound in your next interview.
You need real numbers, including adoption rates, cycle time cuts, cost savings, and revenue. You need artifacts like frameworks, playbooks, dashboards, and guides. You need talent relationships with the eight to twelve engineers who actually get it.
Here is your eighteen month plan. In months one through three, document everything. The tech stack, the organization chart, the politics, and who hates whom. This is your before picture. In months four through nine, pick one team with low political barriers. Embed agents. Eliminate toil. Get them forty percent faster. Measure everything. This is your proof point. In months ten through fifteen, run the same playbook with three more teams. Prove it works across different contexts. In months sixteen through eighteen, decide if you are staying or leaving. By month sixteen you will know if they are serious or just checking boxes.
Consider the market reality. Most companies have no idea what to pay AI leadership. They guess somewhere between two hundred fifty thousand dollars and eight hundred thousand dollars.
Check agent driven development dot com. Look at what is actually possible with agent driven workflows. Then look at Fortune five hundred vice president of AI job descriptions. See the gap.
If you document real outcomes, you become one of maybe two hundred people who can credibly say I have operationalized AI transformation at scale.
Small market. Seller's market.
Companies that understand what they are buying pay multiples more.
Understand what this actually is. You are documenting how AI transformation works in hostile territory. Building frameworks that prove you can execute. Finding talent that ships in chaos. Creating the case study that gets you hired somewhere already serious about this.
You might save this company. But that is not the goal. The goal is getting hired somewhere that does not need saving.
Here is the bottom line. Take the role. Negotiate hard. Execute fast. Document everything.
The market for AI leadership is being written right now. People with real outcomes will define what it pays.
You have got eighteen months.
Make them count.