Which AAES product does what?
| Product | Purpose | Availability and deployment |
|---|---|---|
| AAES Infer | Model access with jurisdiction controls. | Through your own Google Vertex AI or Azure account, under your own provider exemption. Qualification required. |
| AAES Operate | Creates and runs agents, assigns private work, and supports operational reviews and limits on recorded spending. | Client-operated or hosted by AAES. Sold separately or with AAES Govern. |
| AAES Govern | Checks permissions, approvals and spending limits on supported routed actions, then records whether each request was allowed, refused or held for approval. | Client-operated or hosted by AAES. Sold separately or with AAES Operate. |
Operate and Govern are sold separately or together. Installation is assisted. Training and enhanced support are optional and agreed separately. The evaluation package runs in your environment and comes only with a paid, scoped evaluation.
Read the scope and status and execution-path capability matrix for control boundaries. Product facts are also available as JSON; the documentation index links to product, integration and evidence pages.
The problem we solve
AI employees need defined tasks, accountable owners and controlled access. AAES Operate brings agents into your organizational structure and gives their work projects, tasks, skills and connectors. AAES Govern adds ownership, access, guardrails and an audit trail for the actions routed through it.
What AAES Infer offers
Infer provides model access with a chosen processing jurisdiction, through your own Google Vertex AI or Azure account, under terms that prohibit training or secondary reuse of prompts and outputs. Each route requires qualification first. Review billing and availability.
What AAES Operate does
Operate creates and runs agents on your infrastructure or in an AAES-hosted environment, with private projects and tasks, operational reviews and limits on recorded spend. Its controls are separate from Govern. Review Operate’s features, deployment responsibilities and component notices.
What AAES Govern does
Govern checks permissions, approvals and spending before authorizing supported actions routed through it, then records each completed decision. Agents with usable direct credentials can bypass it; observation-only activity cannot be blocked. Record integrity does not prove complete capture or downstream execution. Review Govern’s controls and boundaries.
AAES serves security, identity, platform engineering and risk teams, alongside their existing identity providers, credential stores and monitoring systems.
Contact AAES
Security reports go to security@aaes.ai; everything else to hello@aaes.ai.
Operator status
The aaes.ai website is operated by AAES LLC, a Delaware limited liability company. Registered address for legal notices: 8 The Green STE B, Dover, DE 19901, United States. Contact: support@aaes.ai.
Availability and evaluation
Operate and Govern can run on your infrastructure or be hosted by AAES. Infer runs through your own provider account once its route is qualified.
Start with a free workflow assessment. A paid evaluation is then scoped and priced for your workflow. The evaluation package runs in your environment and comes only with a paid, scoped evaluation. Agree the workflow, support, and acceptance criteria before the evaluation starts.
Public documentation, the Apache-2.0 verifier and sample pack, and published Python and TypeScript SDK packages are available separately. They do not grant a platform software license or access to a running deployment.
No SOC 2 report or independent certification is offered. Model-gateway live results are on the Models page, and reviewed capability packages have completed documentation reviews; live provider-tenant validation is part of a paid evaluation. Read the scope and status, capability matrix, and security statement before evaluating a workflow.
Evidence you can inspect
The public Apache-2.0 aaesverify v0.3.0 source and release binaries verify the aaes.export/v3 exports that current AAES Govern deployments write, and retained aaes.export/v2 exports such as the public sample. The earlier v0.2.0 release reads v2 only.
The offline check uses a public key you trust separately to confirm that exported records haven’t been changed. It can’t prove that every action was captured or that an action actually ran. A production install requires an independent witness or timestamp service, which confirms the signed summary of the record log. Our trust model describes these boundaries and the configuration needed to evaluate them.
How we publish
AAES authors its product documentation and educational guides. Where a page describes another vendor or a governance framework, readers should follow the linked primary sources and the stated review date. Comparisons are written by AAES, which sells a competing product; they are not independent analyst rankings.
We distinguish implemented behavior, documentation review, local tests, and live deployment evidence. A review date records an editorial review, not certification or proof that a customer has accepted a control. Send factual corrections with the page URL and supporting evidence to hello@aaes.ai.
Standards and community.
Working-group participation
Cloud Security Alliance
AAES is a member of the Cloud Security Alliance’s Autonomous Action Runtime Management (AARM) working group.
Explore the AARM working groupContribution proposal
OWASP GenAI Security Project
Our proposal for the Agent Control Standard explores how reviewers can check AI agent action records offline. It is a proposal under public review; follow the discussion for its status.
Read the ACS proposalOfficial links and contact
Our canonical website is aaes.ai. The previous aaes.dev website redirects here.