Human-agent trust for manufacturing & supply chain

Procurement agents close purchase orders at 3am. Automated guided vehicles reroute on AI signals. Customs documents are agent-signed under a human's e-signature. The supply chain audit only works when each step names its human — and the regulators are starting to ask.

The four agent-fragile junctures

Manufacturing audits live or die at four boundaries. Procurement. An agent that opens, awards, and signs a $250k PO is a single approval. Inbound logistics. An agent that accepts a delivery on tonnage and quality grade is the only signature on the receiving doc. Production scheduling. An agent that re-routes work across machines reallocates capital that compliance later has to reconcile. Customs and trade documents. An agent that submits an export declaration under the human exporter's e-signature exposes the exporter to false-statement liability.

What Manav adds, by juncture

Procurement agents present a delegation from the buyer (scope po:create, magnitude cap by category). The buyer's scope encodes the company's signature authority matrix; the agent cannot cut a $1M PO for an $100k scope. Inbound receivers carry a delegation from the warehouse manager; quality flags trigger a multi-signature requirement. Production scheduling agents carry plant-manager scope; capital reallocations above thresholds prompt a co-signature. Customs declarations require a human export officer's signed delegation, separate from the e-signature on the document, with both audit-logged.

The forensics dividend

When a defect, fraud, or counterfeit lands inside a complex supply chain, the audit cost is the cost. The median forensic-trace investigation across a tier-3 manufacturing supply chain takes 18 weeks and costs $1.4M, mostly because the agent layer is unsigned and investigators reconstruct authority by interview. Manav-bound audit trails compress the forensics to days.

What this looks like in ERP

SAP, Oracle Cloud, and Microsoft D365 each accept the Manav delegation as a metadata column on transactional records. Reports filter by delegated-by-human DID. The auditor's quarterly export is a single CSV; the regulator's evidence request is a single signed export. No new system; one new column.

Trade compliance, specifically

OFAC, EU dual-use, and India's SCOMET regimes all penalize false statements on export documents. When an agent files a declaration under a human's e-signature, the human carries the criminal exposure if the declaration is wrong. Manav's pattern requires the agent to present a separate, scoped, time-bound delegation for each filing — turning "the AI did it" into "the export officer authorized this filing in this scope on this date," which is what the regulator wants to read in evidence.

Common objections

The two pushbacks we hear from this vertical: integration risk — addressed by phased rollout starting with the audit trail (lowest risk, highest evidence-to-effort ratio), and internal politics — addressed by anchoring the project to a regulator deadline or a security-questionnaire deal-blocker, where the political question answers itself.

Frequently asked questions

What is the first integration to ship? The signed audit trail. It costs least, satisfies the most regulators, and produces the evidence everything else builds on. Every vertical we have integrated started here.

How does this affect end-customer experience? Invisibly, by design. The customer sees the same UI; the difference is in the audit log behind it. The latency added is single-digit milliseconds. The trust gain is structural.

What's the buying motion — security, compliance, or the line? Compliance writes the check; security signs off; the line of business sets the timeline. The strongest deals start with a regulator deadline; the next-strongest start with a deal-blocking security questionnaire.

Where to start

The first integration we recommend in this vertical: agent identity finance, then audit trail design. Both are deployable inside a quarter; both produce regulator-grade evidence; both unblock procurement conversations the rest of the stack depends on.

Adjacent reading

For the regulatory ground truth in this vertical, see the Article 14 playbook. For the integration shape, see audit-trail design. For the buying motion, see the CISO compliance stack. Most successful pilots in this vertical follow that order: regulation first, integration second, procurement third.

Why supply-chain auditors will eventually demand this

Supply-chain auditors are roughly two years behind financial auditors in their use of cryptographic evidence, and the gap is closing. The convergence is being driven by counterfeit incidents, recall investigations, and trade-compliance enforcement actions where the agent layer was the unsigned link the investigators could not bridge. Each significant incident accelerates the convergence. The next major recall investigation that names a delegation gap as a root cause will move supply-chain auditors from informal expectation to formal requirement, and we expect the move to happen within the regulatory cycle currently in flight. Manufacturers ahead of the move ship audit-grade evidence on day one of the requirement. Manufacturers behind it ship under deadline at substantial premium cost. The procurement decision is therefore less about whether the requirement arrives and more about which side of the announcement you are positioned on when it does. We are advising customers as if the announcement is already in the regulatory queue, because in our reading it is.

An audit trail that names "the system" is not an audit trail. It is a polite fiction.