// AI Advantage
The One-Page Policy That Stops Runaway AI Spend
The prompt that drafts your AI Token Spend Policy — caps, dashboard, and agent tiers — in one paste
One enterprise reportedly spent half a billion dollars on Claude in a single month after rolling it out to thousands of employees with no usage caps. Nobody got hacked — nobody watched the meter. I keep hearing softer versions of the same story from operators every week, and Gartner's read on why agent programs fail is blunt: it's governance, not capability. The fix isn't a committee or a quarter-long project. It's a one-page policy with three controls, and the prompt below drafts it for your company in about two minutes.
Paste this into Claude or ChatGPT, fill the brackets, and send the output to your leadership channel today:
You are an AI FinOps lead. Draft a one-page AI Token Spend Policy for
[COMPANY], a [SIZE AND INDUSTRY] company spending roughly [$X/MONTH] on
AI across [TOOLS IN USE, e.g., Claude, ChatGPT, Copilot, API access].
The policy must ship exactly three controls:
1. SPEND CAPS — Per-user and per-team monthly caps on every tool with
usage-based pricing. Default cap: [$X PER USER]. Include an exception
path for high-value workflows, owned by [ROLE], so the policy funds
wins instead of just blocking spend. Alert at 80% of cap.
2. VISIBILITY — A daily-refreshed dashboard of AI spend by team, tool,
and model, owned by [ROLE], with an alert to [CHANNEL] whenever a team
runs 20% above its trailing 30-day average.
3. AGENT TIERS — A written rule for which agentic workflows run
unattended and which need a human checkpoint. Tier 1: low-risk, runs
unattended. Tier 2: human reviews output. Tier 3: human approves before
any external action or spend above [$X]. Tier by risk, not uniformly.
Format: one page, plain language, an effective date, and a 90-day
review. Close with the question every team lead answers monthly:
"What was our value per token this month?"
Sample output (excerpt from the caps section, run for a 400-person software firm):
Spend Caps. Each user receives a $150/month default allocation across
Claude and Copilot. Team caps equal headcount × $150. Requests to
exceed a cap go to the VP of Engineering with a one-line business
case; approved exceptions are logged and reviewed at 90 days. Alerts
fire at 80% of any cap — no one learns about a limit by hitting it.
What changes: You stop discovering AI spend on the invoice and start governing it like any other metered utility. The cap is the circuit breaker, the dashboard is the meter, and the tiers keep your highest-value agents running while the risky ones get a human in the loop.
Where else this works: The same skeleton drafts a cloud egress policy, a SaaS seat-audit policy, a vendor API budget, or an experimentation budget for a data team — anything metered, variable, and invisible until the bill lands.
The timing matters. Forbes reported this week that enterprise agent spend keeps climbing even as providers cut token prices, and CIOs now rank token spend as the metric to watch — with 98% of organizations actively managing AI spend, up from 31% two years ago. The companies getting this right didn't buy new software first. They wrote the page. Draft yours before the next invoice — it's ten minutes, and it's the difference between doing productivity math and doing damage control.
