TenGrid Security · AI Risk Management
AI Risk Assessment
A one-time assessment that maps your actual AI/ML footprint — which models you run, where they came from, and where they are exposed — and turns it into a prioritized list of what to fix first.
$3,997 one-time
Secure checkout by Stripe · We email you after checkout to schedule scoping
Who it’s for
Organizations running ML models or AI features in production, or adopting third-party and open-source models, that do not yet have a clear picture of their AI/ML risk.
What’s included
- AI/ML footprint inventory — the models in scope, their owners, third-party vs. in-house, and documented data sources
- Model integrity & supply-chain review — in-scope model files and artifacts scanned for tampering or unsafe content
- Adversarial exposure review — priority model endpoints reviewed for adversarial input and prompt-injection exposure, rate limiting, and access control
- Prioritized findings + remediation roadmap — what to fix first, and why
- Readout call — walk through the findings with your team
How it works
Pay securely through Stripe.
We agree which models and endpoints are in scope.
Delivered by our security team.
Prioritized report and roadmap, reviewed with your team.
Questions
Who performs the assessment?
TenGrid Security scopes and manages the engagement. The technical assessment is delivered by our security team.
Is this a compliance audit?
No. It is a risk assessment of your AI/ML footprint. It is not a compliance determination or certification.
What do you need from us?
A point of contact and access to the in-scope models, artifacts, or endpoints agreed on the scoping call.
What comes after the assessment?
Ongoing coverage is available under Managed AI Risk Monitoring ($1,497/month).
Map your AI/ML footprint
Related
Managed AI Risk Monitoring
$1,497/month
Ongoing monitoring of your in-scope models and model endpoints, delivered by our security team.
View managed monitoringAI Risk Exposure Report
$300 one-time
Where AI/ML risk actually shows up, and what changes when it's managed. Instant PDF after checkout.
See the reportEducational content and service descriptions reflect general industry practice. Nothing here is a security audit, a compliance determination, or a guarantee of any outcome. Model and AI system risk varies by deployment.