Optimize AI output IP risk portfolio

What it's for

Shows where scanning, licensing, redesign, replacement, exclusion or insurance creates the best risk-adjusted AI-output value—and when apparent indemnity is not collectible.

What you give it

Inputs split into evidence read from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.

{
  "counterparties": [
    {
      "id": "provider-a-indemnity",
      "recovery_available_probability_scenarios": [
        0.95,
        0.7
      ]
    }
  ],
  "implementation_budget": 100,
  "output_classes": [
    {
      "artifact_count": 20,
      "business_value_scenarios": [
        50000,
        50000
      ],
      "claim_prior_alpha": 2,
      "claim_prior_beta": 18,
      "counterparty_group_id": "provider-a-indemnity",
      "delay_value_per_hour": 50,
      "false_positive_cost": 100,
      "id": "code-output",
      "loss_per_claim_scenarios": [
        20000,
        100000
      ],
      "maximum_claim_probability": 1,
      "provider_group_id": "provider-a",
      "required_control_ids": [
        "counsel-review"
      ]
    }
  ],
  "provider_groups": [
    {
      "id": "provider-a",
      "unique_value_at_risk_scenarios": [
        50000,
        200000
      ]
    }
  ],

What you get back

This is the actual output of running the example above — computed by the same function the platform calls, not an illustration.

{
  "assumptions": [
    "Claim prevalence and locally validated action effects are exchangeable within each aggregate output class conditional on represented scenarios; shared provider claims and counterparty recovery availability remain correlated across the portfolio.",
    "Business value and common provider value are unique, costs and capacity are complete, indemnity limits/deductibles are enforceable under counsel-approved terms, and every executable option is represented before optimization."
  ],
  "baseline_current_state": {
    "expected_net_value": -58670.95,
    "expected_total_loss": 108654.2,
    "loss_conditional_value_at_risk": 1065016,
    "selected_option_ids": [
      "keep"
    ]
  },
  "class_diagnostics": [
    {
      "action_type": "similarity_scan",
      "claim_probability_by_scenario": [
        0.0271,
        0.0993
      ],
      "expected_escaped_claims": 0.042,
      "expected_net_direct_loss": 2668.4,
      "output_class_id": "code-output",
      "selected_option_id": "scan-and-remediate"
    }
  ],
  "decision": "implement_selected_ai_output_ip_risk_portfolio",
  "economics": {
    "expected_common_provider_loss": 2675,
    "expected_gross_direct_claim_loss": 4530,
    "expected_indemnity_recovery": 1861.6,
    "expected_net_direct_claim_loss": 2668.4,
    "expected_net_value": 43031.9,
    "expected_operating_cost": 524.7,
    "expected_retained_business_value": 49000,
    "expected_total_loss": 5343.4,
    "implementation_cost": 100,
    "loss_conditional_value_at_risk": 107398,
    "loss_value_at_risk": 800,
    "risk_adjusted_score": 16182.4,
    "tail_probability": 0.05
  },
  "limitations": [
    "Optimization ranks a declared internal planning model; it is not legal advice, a coverage opinion, an infringement finding, procurement authority or permission to publish or commercialize an artifact.",

How it works

Constrained optimization — Pick the best feasible option under real limits — budget, headcount, dependencies, capacity — rather than ranking a list and hoping it fits.

1. 1 Draw class claim prevalence and coherent scenarios, then simulate each option's claim reduction, detection, remediation, false positives, delay, gross loss and collectible indemnity under shared counterparty availability.
2. 2 Count each common provider's unique value once through joint detection survival while preserving common claim states across every dependent output class.
3. 3 Reject portfolios that breach claim, evidence, control, dependency, exclusion, budget or capacity gates; maximize expected net value minus loss CVaR by exact enumeration or a disclosed deterministic beam and expose the Pareto frontier.

Before you trust it

Every tool in the catalog ships with the conditions under which its answer is meaningful — and the conditions under which it should abstain instead of guessing.

Calibrating it to your org

Calibration workflow

1. 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
2. 2 Build a tenant-scoped historical cohort using only information available before each prediction or decision; preserve zero periods, censoring, assignment probabilities, and unresolved outcomes when the method requires them.
3. 3 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
4. 4 Validate on later time windows or held-out aggregate units at the deployment grain, against a simple baseline and the function-specific validation strategy.
5. 5 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
6. 6 Monitor realized outcomes, data drift, coverage, and decision regret; recalibrate at a governed cadence or after a detected regime/definition change, never merely because a stakeholder dislikes the result.

Call it from your AI

You don't wire up 388 tools in your MCP client. The GitRevio MCP server exposes 18 tools, three of which let an agent search the catalog, read a tool's schema, and run it — so the assistant finds this one on its own.

gitrevio_capabilities_search
  { "q": "choose keep scan license redesign replace" }
  → finds "optimize_ai_output_ip_risk_portfolio"

gitrevio_capability_describe
  { "capability_id": "optimize_ai_output_ip_risk_portfolio" }
  → returns the input schema and agent guidance shown on this page

gitrevio_capability_run
  { "capability_id": "optimize_ai_output_ip_risk_portfolio", "arguments": { ... } }
  → returns the result shown above

Works in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.

Related tools

See every tool in AI risk, rights & assurance →