AI-generatedPriya Raman
- Actor role
- developer, deployer
- Risk tier
- medium
- Confidence
- 100 · High
- Lifecycle
- production
As a developer, deployer on a medium-risk production, your self-attested readiness is "Substantially ready" (84/100). You are strongest in map — context, intended purpose & impact profile; the most material gap is in measure — tevv, metrics & trustworthiness evaluation.
Your results are available through this secure link.
Confidence adjusts 84 → 84.
Reflects how complete and consistent your profile and current-state answers are. It never raises readiness.
Foundational gaps that cap the score and cannot be averaged away. See findings below.
What this assessment indicates
As a developer, deployer on a medium-risk production, your self-attested readiness is "Substantially ready" (84/100). You are strongest in map — context, intended purpose & impact profile; the most material gap is in measure — tevv, metrics & trustworthiness evaluation.
Function performance
Score out of 100 · target 70Trustworthiness overlay
A secondary view of how your implemented controls map to NIST’s seven trustworthiness characteristics. It does not double-count into the four function scores.
Top strengths
- Are AI risk management policies, processes and procedures documented and implemented across teams? GOVERN 1
- Does the organization determine the needed level of AI risk management based on risk tolerance and context? GOVERN 1.3
- Is there a resourced inventory mechanism for AI systems, components and workflows? GOVERN 1.6
Top gaps
- Are GAI risk tiers defined for information integrity, harmful content, rights impacts, security vulnerabilities and variability over time? GV-1.3-001
- Are minimum performance/assurance thresholds used in deployment approval? GV-1.3-002
- Does the system handle malicious or illegal requests such as manipulation, extortion, cyber-attacks or weapons creation? MS-2.6-006
- Are GAI vulnerabilities such as prompt injection, model extraction and data poisoning tested? MS-2.7-007
- Is there an inventory of data sources indexed or retrieved by the system? MAP / GOVERN
Prioritized remediation roadmap
- P0Implement GenAI content safety, provenance and misuse controlsAI 600-1Owner: Product / SecurityBefore launch/continued use / 0–30 days
- P0Test GAI vulnerabilities: prompt injection, extraction, poisoningMS-2.7-007Owner: SecurityBefore launch/continued use / 0–30 days
- P0Secure and monitor RAG data sources, permissions and grounding qualityAI 600-1 / MAPOwner: Data / PlatformBefore launch/continued use / 0–30 days
Turn these findings into a remediation plan
Unlock the detailed report for a function-by-function gap matrix and a 30/60/90-day roadmap, or request a verified review with evidence and analyst input.
This is an informational, self-attested readiness result — not a NIST endorsement, certification, audit, conformity assessment, or proof that an AI system is safe, fair, valid or trustworthy. No documents or evidence were reviewed. Foundational caps prevent a strong area from hiding a missing foundation; the confidence score reflects how complete and consistent your answers are, not whether controls truly exist.