PAGE 05 · THE GAP

Pilot AI Agents Safely

An agent that can use tools expands both capability and failure impact. Begin read-only, limit every permission, and require human approval before money, messages, records, or customer commitments can change.

Least privilege · evaluation · approval · rollback
Educational architecture, not deployment advice. Vendor features, regulations, security requirements, costs, and appropriate controls vary. A working demo is not evidence that a system is safe, compliant, profitable, or ready for production.

Decide whether an agent is necessary

Prefer a deterministic workflow when the steps and rules are known. Consider model-assisted planning only when variability justifies it and the result can be checked. Never give an agent authority merely because it produces convincing text.

Task: the one job it may attempt

Tools: an explicit allowlist with narrow scopes

Data: permitted fields, tenants, purpose, and retention

Approval: actions a qualified human must confirm

Limits: spend, recipients, records, rate, time, and retries

Recovery: pause switch, idempotency, fallback, and rollback

Enforce controls outside the prompt

Authentication, authorization, tenant isolation, data validation, tool allowlists, spend caps, recipient limits, idempotency, rate limits, and approval must be enforced by trusted services. Treat model text and retrieved content as untrusted input. Never place credentials in prompts, browser code, logs, or lesson screenshots.

Evaluate before action

  1. Create a versioned set of representative, multilingual, adversarial, and edge cases.
  2. Measure task completion, false actions, omissions, reviewer corrections, latency, and cost.
  3. Test prompt injection, privilege escalation, data leakage, duplicate action, and tool outage.
  4. Run offline, then in shadow mode, then read-only with a limited consenting group.
  5. Define task-specific release and stop thresholds before seeing the result.
  6. Regression-test every model, prompt, tool, permission, or policy change.

Commercial and privacy boundaries

The contract should define data roles, authorized environments, subprocessors, retention, IP, confidentiality, service levels, human responsibilities, incident handling, change control, and liability. Do not mix customer data, evaluation sets, prompts, or outputs across clients without explicit written rights.

If the agent communicates externally, identify the organization, disclose automation where appropriate, use only permissioned recipients, honor opt-outs, and prohibit unsolicited bulk contact. No agent should impersonate a person or hide AI use to win trust.

Use evidence, not projected savings

ZERO-DEFAULT PILOT SHEET

Eligible volume (0), baseline effort (0), reviewed pilot effort (0), approved cost basis (0), error impact (0), tool and oversight cost (0), incident cost (0). Any calculated benefit is hypothetical until repeated customer measurements support it; include negative and inconclusive outcomes.

Completion checklist

Increase authority only after evidence. Keep consequential actions reviewable.

A responsible pilot proves that controls work under failure—not only that the happy path can complete.

Built by Avamartech · Production use requires independent review