THE OFFER

The AI Production
Risk Audit.

One fixed-scope engagement. Four risk surfaces, scored independently and ranked by business impact. Delivered as a single document, not a slide deck.


WHO IT'S FOR

Built for teams that shipped fast and never went back to check.

Series A–C tech companies, typically 20–300 people, who have AI features in production but no one formally responsible for verifying they're safe at scale. Usually triggered by a near-miss, a board question, a security questionnaire from an enterprise customer, or a new technical leader inheriting a system they didn't build.


WHAT GETS CHECKED

Four risk surfaces. Every system, every time.

The same framework applied consistently, so findings are comparable and nothing gets missed because it wasn't the obvious place to look.

01

Data exposure

What the AI layer can read from and write to, whether access matches what it actually needs, whether permissions have drifted since launch, and whether there's any audit trail for what it accessed and when.

02

Architecture brittleness

Single points of failure, vendor lock-in, undocumented dependencies, and bus factor — how many people understand each critical piece, and what breaks if one of them leaves.

03

Output reliability

Where AI output reaches a customer or a high-stakes decision without review, what the blast radius is if it's wrong, and whether error rates are measured or just assumed to be fine.

04

Operational readiness

Headroom at 5–10x current usage, clear ownership of the system today, recovery time if it fails, and whether there's a rollback path if a change makes things worse.


SAMPLE OUTPUT

What the report actually looks like.

A synthetic example, run against a typical AI support-ticket assistant. Real engagements follow this exact structure — ranked findings, business impact first, technical detail second.

sample-audit-support-assistant.report
SYNTHETIC EXAMPLE
2
critical findings
2
high findings
1
watch findings
14 days
audit length

Standing read access to the full customer table

5 × 4 = 20Critical
DATA EXPOSURE

The support assistant has unrestricted read access to the customer database, not just the fields it needs to answer tickets. Access was never scoped down after the prototype stage, and no one currently owns reviewing it.

Recommended fix: Scope the service account to a read-only view of ticket-relevant fields; add quarterly access review.

No rollback path for prompt changes

4 × 4 = 16Critical
OPERATIONAL READINESS

A prompt update that degraded answer quality was live for nine days before anyone noticed — there was no versioning or fast revert path, and no monitoring flagged the drop in resolution rate.

Recommended fix: Version prompts in source control; add a same-day rollback path and a weekly resolution-rate check.

Single undocumented retrieval script

3 × 4 = 12High
ARCHITECTURE BRITTLENESS

The entire retrieval pipeline depends on one script written by a former contractor. No one currently on the team has modified it or could explain several of its assumptions.

Recommended fix: Document the script's logic and dependencies; assign a current owner before any further changes.

No measured error rate on generated responses

4 × 3 = 12High
OUTPUT RELIABILITY

Generated responses reach customers with no human review step and no logged accuracy rate. The team learned about a wrong pricing answer from a customer complaint, not from monitoring.

Recommended fix: Sample and score 5% of responses weekly; route low-confidence answers to human review.

Cost scales unpredictably under load

2 × 3 = 6Watch
OPERATIONAL READINESS

Per-ticket model cost is not currently tracked or capped, so a spike in ticket volume would translate directly into an unbudgeted cost spike with no early warning.

Recommended fix: Add per-day cost tracking and an alert threshold; revisit at next audit.

WHAT HAPPENS NEXT

Critical and High findings become your remediation roadmap.

The audit report is the spec. You decide which findings to act on and in what order — Strataforge3 can scope and quote fixes individually, with no open-ended retainer required.

Know exactly where the risk is before someone else finds it.

A 20-minute call to see if an audit makes sense for what you've built.