AI Governance
Route: /org/resolve/ai-governance · Permission: AI Governance
1. What it is
What the AI features actually cost and how reliably they run.
2. What is measured
| Panel | Shows |
|---|---|
| Daily AI calls | Volume over time |
| By model | Which model consumed what |
| By utility | Which feature consumed what |
| Calls · Credits · Tokens · Failed | The four totals |
Failed calls still consume
A rising Failed count alongside steady credit spend usually means a misconfigured utility retrying. Compare Failed against Calls regularly — it is much cheaper to notice here than on an invoice.
"By utility" tells you which feature to tune
Spend concentrated in one utility is the place to look first — a prompt that could be shorter, or a generator being used where a template would do.
3. Don't confuse this with…
| These two | Differ by |
|---|---|
| AI Governance vs Integration Telemetry | AI spend vs API traffic |
| AI Governance vs AI in Resolve | What it cost vs what the features do |
| Credits vs Tokens | Billing unit vs model unit |
4. Troubleshooting
| Symptom | Cause |
|---|---|
| Spend jumped | Check By utility for the feature responsible |
| High Failed count | A misconfigured utility retrying |
| No data | No AI features used in the period |