Disclosures
The fine print, said plainly. We'd rather you read this now than discover it in a contract later.
1. Some of this is live. Some is active R&D.
We are a research lab as much as a product company. Parts of what these docs describe run in production today; others are under active research and development, and capabilities land at different times. The Status board is the single source of truth, and before you commit to anything we'll tell you exactly which pieces are ready for your use case and which are still on the bench.
2. Configured with you, not off the shelf.
Deployments are assembled per customer and validated to your satisfaction before go-live. That's the point of the compiler. It also means we do not promise that every capability described in these docs exists as a general, ready-made feature for every workload. What we deliver for you is what we agree with you, in writing. Your agreement with zkAGI defines the guarantees, availability and service levels that apply, and that agreement is what governs, over anything on this site.
3. Your audit stays your audit.
zkAGI is not an audit firm, and our inference service has not itself been audited or certified against HIPAA, GDPR, the EU AI Act, or any similar framework. Compliance responsibility stays with you, the customer. Our job is to make that job easier: receipts, attestations and documentation built to be handed to your auditor, plus hands-on help getting ready. We assist with audit preparation; we do not conduct audits, issue certifications, or give legal advice. Regulatory mappings in these docs (for example, the table on the Receipts page) are orientation guidance for your compliance team, nothing more.
4. Benchmarks are ours unless stated otherwise.
Figures like token-recovery bounds are produced by our own benchmark suites under stated conditions. They are honest, repeatable, and not a substitute for third-party validation. Where a specific deployment needs independent measurement, we'll support it.
5. Peaceful use only.
zkAGI is built for peaceful, civilian use: business productivity in regulated industries. We do not take defense or other dual-use work.
6. Trademarks and names.
NVIDIA, Intel TDX, AWS Nitro, Azure Confidential and similar names are trademarks of their respective owners, referenced for comparison and compatibility description only. OpenAI compatibility refers to API-shape compatibility with the OpenAI client SDKs, and implies no affiliation.
Questions about any of this deserve a human answer: talk to us. Nothing on this page or this site is legal advice.