AI Acceptable Use Policy + Agent Commerce Guardrails (L5)

Two open specs to bring sanity to org-readiness: an AI AUP baseline that combats shadow AI without over-restricting, and Agent Commerce guardrails that define what terms/$ limits an agent may auto-approve. From John Williams (FXOps) on the GTM AI Podcast.

WORKFLOW1Adopt the open AI AUP as …our v1 baselinegithub.com/fxops-ai2Define agent spend + ToS …imitsAgent Commerce spec3Audit existing agents aga…nst both specsManual4Publish the spec internal…y + bake into procurementInternal wiki
4 steps, in order, with the tool that owns each one.
Adoption ladderSix levels from Starter to Rebuilt. This item sits at level 5.L1 StarterOne tool, no workflow changeL2 AssistedAI drafts, humans approveL3 IntegratedWired into CRM and SlackL4 OrchestratedMulti-step, owned, measuredL5 AutonomousAgent runs, human auditsL6 RebuiltThe process itself changes
This playbook belongs at L5 Autonomous. Running it above your level is how pilots stall.
Measures of success% of agents AUP-attested; # of shadow-AI incidents/quarter; time-to-onboard a new AI tool; # of unauthorized agent transactionsPROVE IT WORKED% of agents AUP-attested# of shadow-AI incidents/quartertime-to-onboard a new AI tool# of unauthorized agenttransactions

The steps

  1. 01

    Adopt the open AI AUP as your v1 baseline

    Tool: github.com/fxops-ai

    Don't start from zero. The open AUP gives Enablement/HR/Legal a workable baseline they can adapt, beats six months of legal review while shadow AI runs the company.

  2. 02

    Define agent spend + ToS limits

    Tool: Agent Commerce spec

    For every autonomous agent: max $ per transaction without human approval, allow-list of ToS your agent may auto-accept, escalation path. Codified in a machine-readable spec the agent checks itself against.

  3. 03

    Audit existing agents against both specs

    Tool: Manual

    Inventory every agent (yours + vendor-deployed). Score each on AUP compliance + Agent Commerce fitness. Anything failing either gets paused until remediated.

  4. 04

    Publish the spec internally + bake into procurement

    Tool: Internal wiki

    New AI tool intake form requires AUP + Agent Commerce attestation. Vendors that can't answer don't get bought.

Tools in this playbook

  • github.com/fxops-ai
  • Agent Commerce spec
  • Manual
  • Internal wiki

Next playbooks

Unfamiliar terms are defined in the AI and Revenue Dictionary. Related frameworks live in the framework library.

Share this playbook

Posting to Instagram or TikTok? Copy the link, it carries the title, summary and share image.

Arrives weekly by email. Free. Unsubscribe anytime. By subscribing you agree to our Privacy policy and Terms. We never sell or share the list.