Adoption outran governance
AI got adopted before anyone decided anything.
Client-facing teams are already pasting work into AI tools.
AI rules / 06
AI arrived before the rules
Adoption signal
Teams are already using it. Leadership owes rules and results.
Recommended service
AI governance & readinessContext behind the pressure
Two departments are running pilots that do not know about each other. The board wants to know what AI is doing for the business; legal wants to know what data is leaving it. You are being asked for a strategy and a policy at the same time—and the only two moves that are obviously wrong are banning everything and blessing everything.
You are likely here if
The pressure is specific. The answer is not yet.
- 01
You cannot list the AI tools in use today, what data goes into them, or which ones touch customer information.
- 02
There are pilots and enthusiasm, but no one can name a workflow where AI measurably changed cost, speed, or quality.
- 03
The board or leadership asked for “our AI position,” and the current answer is a policy draft in one tab and a vendor demo in the other.
The decision on the table
Name the decision before choosing the deliverable. It keeps the work tied to the business question that created the pressure.
Make two decisions together. The governance decision: what data may go into which tools, under whose oversight. The investment decision: which single workflow gets a real pilot with an owner, a baseline, and pre-agreed scale, revise, or stop criteria. Answer only the first and you become the department of no. Answer only the second and exposure scales alongside output.
Ackatec’s working view
A position before a proposal.
- 01
Ground rules beat policy documents.
Two pages people can follow this week—off-limits data, sanctioned tools, human-review points, an exception path—do more than a policy ignored in practice.
- 02
One governed workflow beats five pilots.
A single workflow with a baseline, an owner, and human oversight either earns its scale decision or gets honestly stopped.
- 03
The risk map and the value map are the same map.
The data-handling picture needed for guardrails is the same picture that shows where AI can operate. Building them separately repeats the work.
A representative first artifact
AI ground rules and first-workflow brief
Replace “we should do something about AI” with a named owner, one workflow, and a date.
Working structure
- 01
Usage inventory
The tools actually in use, who uses them, and what classes of data they touch.
- 02
Ground rules
Sanctioned tools, off-limits data, human-review points, and the exception path—short enough to be followed.
- 03
Workflow shortlist
The workflows where AI plausibly changes a number the business tracks, and the one to pilot first.
- 04
Pilot frame
Owner, baseline, guardrails, oversight, and the pre-agreed scale, revise, or stop criteria.
- 05
Decision calendar
When leadership reviews the evidence and makes the call.
The ground rules ship first and buy time; the pilot frame turns the leadership question into an evidence-based answer by a known date.
This is Ackatec’s working structure for the situation—shaped to each organization in the first sessions, not a sample of past client work.
Recommended next step
Get the right help for this moment.
Begin from the situation