For Engineering

GenZAgents for product managers — ship AI features your customers can audit

Customers buying AI-powered products increasingly want to audit them. Your "AI summary" / "AI search" / "AI agent" features need a receipt layer. GenZAgents ships the trust artifact your customer experiences as transparency.

Why customers want to audit AI features

For consumer products: regulatory transparency (EU AI Act §50 transparency). For B2B products: their own audit obligations (their SOC 2 audit asks "what AI is in your vendor stack and what does it do with my data?"). Without an audit layer, your customers can't answer these questions and they'll pressure you to add one or push you out for a vendor that has one.

How to ship "trust as a feature"

For each AI-mediated action in your product, issue a receipt via the GenZAgents SDK. The receipt captures the model used, the customer's identity, the inputs (digest), the outputs (digest), the cost. Expose the receipt ID in the UI: "AI Summary — receipt: rcpt_01XK… Verify". Customers click through to verify the receipt independently. Trust is now a one-click-from-the-UI feature.

Per-customer audit panel

Within your product's settings: an "AI activity" panel that lists every AI-mediated action this customer triggered, with date, action type, model, cost, receipt ID. Backed by /v1/receipts queries scoped to the customer's identity. This is the auditability customers want; you ship it as a few hundred lines of glue code over the GenZAgents SDK.

White-label option

Enterprise tier ships white-label receipt issuance: receipts are signed by your domain (not by genzagents.com). Customers verify against your /did.json instead of ours. The audit pattern is the same; the brand is yours. Useful when you want to ship trust without depending on a third-party brand.

Use case: AI search in a SaaS product

Your product has AI-powered search. Every search query becomes a receipt: customer ID, query, top results returned (digest), model used (Claude / OpenAI), cost. Surfaced in the customer's audit panel as a queryable feed. Their compliance team can query "show me every AI search that returned data flagged as confidential". This is the level of transparency that wins enterprise sales today.

Use case: AI agent in a workflow product

Your product runs autonomous agents that perform multi-step work for the customer. Each run is a receipt — what the agent did, which tools it called, what it produced, the cost. Customer's audit panel shows the agent's work history; their dispute mechanism uses the receipt as the audit anchor. Drives conversions on enterprise contracts.

Common questions

Can I integrate this with my existing product UI?

Yes — the SDK is a few function calls. We ship React components for common patterns (audit panel, receipt verifier widget). Drop in or copy + customise.

What's the impact on my pricing model?

You'd typically add audit trail as a feature of higher tiers or as a separate compliance add-on. Most products price it at 10-20% of base subscription; some make it free as a sales tool.

Can my customers verify receipts without using GenZAgents themselves?

Yes — receipts are publicly verifiable using standard Ed25519 + JCS tooling. We provide a verifier at /verify; offline verification works with just openssl + jq + a Node script.

Does this work with our existing observability (LangSmith, Datadog)?

Complementary. LangSmith / Datadog are for you; GenZAgents receipts are for your customers. Most teams use both.

Related

Get the trust layer for your AI work

GenZAgents is the verified work-history layer above every AI provider your team uses. Sign cryptographic receipts, hand off conversations across Claude / ChatGPT / Cursor / Gemini, keep institutional AI knowledge when employees leave.

Last reviewed · 3 min read· Open spec· Changelog