MENUCLOSE
Honest coverage
What Tallin governs, and what it doesn't.
Each person, application, and autonomous agent carries one of four coverage labels: observed, inferred, enforced, or not covered. Tallin governs the AI routed through it, and tells you plainly what falls outside. No false certainty.
Coverage taxonomy
One honest language for the state of every AI actor.
The four states keep finance, security, legal, and platform teams reading the same map: which AI actors Tallin is actively governing, which it can only observe, and where the real gaps remain.
- OBSERVED
- Observed: directly captured through gateway events, provider usage APIs, provider admin APIs, or authoritative telemetry, attributed to the AI actor behind it, whether a human, an app, or an agent.
- INFERRED
- Inferred: likely usage from billing, expense, SSO, network, CSV, or discovery evidence.
- ENFORCED
- Enforced: actively governed in the request path (an owner, an approved model scope, a budget with an automatic cutoff, a policy, and an off switch for the actor) plus access restriction, SSO, CASB, or provider admin policy.
- NOT COVERED
- Not covered: anything outside the AI Tallin routes: direct provider calls made outside Tallin, and tool / MCP execution side effects. We report these as explicit gaps, never silenced, so the picture stays honest.
Why it matters
Autonomous agents are arriving fast. Gartner projects 150,000+ AI agents per Fortune 500 by 2028, yet only 13% of teams feel adequately governed. They spend and act unattended, so a dashboard that quietly blends enforced controls with inferred guesses is dangerous. Tallin keeps the four states visible so leaders can act on what they govern without pretending the picture is complete.
Sources grouped by how buyers think.
no endpoint agents to installJoin the design partner cohort.
Tell us about your AI actors, provider stack, and the runaway spend or policy gaps you need to close. We will help you decide whether Tallin is a fit for the first cohort.