Who is the x402 Foundation's executive director? Michael Hursta — and the authorization layer the standard still needs

On October 1, 2026, the Linux Foundation appointed payments industry veteran Michael Hursta as executive director of the x402 Foundation — the 50+ member organization writing the neutral open standard for internet-native payments across AI agents, applications, and APIs. Hursta's résumé is the story: head of payment acceptance and experience for North America at Amazon, leader of the enterprise payments advisory practice at PwC, partner at Commerce Technology Partners, more than a decade at First Data. As Linux Foundation CEO Jim Zemlin put it, his leadership "bringing new payment technologies from emerging concepts to commercial scale will be invaluable as this community moves that work forward." That is exactly the right person to commercialize a machine-payments rail. It is also exactly the moment to ask the question the appointment announcement never touches: a standard for machine payments with no authorization layer is a settlement rail, not commerce. The machine-native answer is the decision gate — ≥0.80 auto-pay, 0.50–0.79 human confirm, <0.50 block, log, and escalate — codified in the standard itself, not bolted on by each implementer.

What actually happened (receipts, dated)

What the field lacks (every one of them)

The October 1–2 field is announcement coverage: the Linux Foundation press release (the primary source — appointments and mandate, zero authorization content), advfn and the lifestyle-wire reprints (pure syndication), briefglance (the strongest narrative take — "cash registers for machines" — but still zero on who authorizes a payment), citybiz (rewrite). Every one of them answers who runs the Foundation. None of them asks the standards question: when 50+ organizations write the rules for machine-to-machine payments, where does the per-payment judgment live — and who records it? That gap is this page's whole reason to exist: a standard that certifies settlement without certifying judgment certifies the exact failure mode behind the $78,000 OpenAI Codex runaway (our breakdown), now conformant by design.

The standards gap: settlement is specified, authorization isn't

The design thesis: a payments standard for machines must score before it settles. Settlement says the signature verified; judgment says the payment deserves to exist. The standard the x402 Foundation writes should require every conformant payment to carry an authorization record — a confidence score, the band that fired (≥0.80 auto / 0.50–0.79 confirm / <0.50 block+log+escalate), and the authority chain behind it. That record is also exactly what the FTC's Sept 30 probe demands of agent builders. Bake it into the standard and the whole ecosystem inherits compliance-grade authorization for free; leave it out and every member reimplements it alone — or never. The pattern we plant our flag on.

SML products that implement the missing layer

Two of the seven ScriptMasterLabs products ship the authorization layer this page argues the standard needs:

• Decision-gated payments — the canonical pattern: every machine payment carries a confidence score, and the band decides — ≥0.80 auto-pay, 0.50–0.79 human confirm, below 0.50 block, log, escalate. The live gate endpoint used for today's receipts (/api/harness/decide) runs this pattern now.
• Provider Trust — trust scoring for the payees themselves: which endpoints, merchants, and counterparties deserve your agent's money. Authorization has two halves — should this payment fire (the gate) and does this payee deserve it (trust). A standard needs both.

Live gate receipts (minted today, not hypotheticals)

Scored against the live endpoint POST https://scriptmasterlabs.com/api/harness/decide — local-heuristic-v1, gate bands 0.80/0.50 live on the endpoint, October 3, 2026 ~09:40 EDT. Exact shape: {"state": {...}, "questions":[{"id","type":"score","scale":[0,1],"question"}]}.

Honest finding: the uncalibrated heuristic scores both questions identically (0.35 → escalate) and cannot discriminate the proposal from the danger. Fail-closed ceiling is the protection — nothing auto-pays, neither the standard change nor the unjudged certifier. That's the seventh consecutive run with this finding (seatbelt, XRPL, Sigo, Robinhood, zkAPI, Cloudflare, now this page): the heuristic is a fail-closed floor, and a production standard needs a calibrated decider behind it — which is exactly the JEV-27B-class slot the Foundation's conformance program should fill. Meta: decider=local-heuristic-v1, calibrated=false, typesafe_wired=false.

Do it yourself: ship the authorization layer the standard doesn't yet require, in 5 steps

1. Score before you sign. Treat every payment authorization your agent issues as a scored decision, not a boolean wallet check. The exact live shape is above — curl -X POST https://scriptmasterlabs.com/api/harness/decide with {"state":{...},"questions":[{"id","type":"score","scale":[0,1],"question"}]}. Bands: ≥0.80 auto, 0.50–0.79 human confirm, <0.50 block, log, escalate.
2. Log like the FTC is watching. Record instruction, authority, intent, and outcome for every agent payment — the September 30 FTC rogue-agent probe (Anthropic, OpenAI, METR) and the September 22 six-bank audit demand made that the regulator's ask, not ours. A standard without a conformance-level audit-trail field forces every member to invent one; invent yours now and you'll be spec-ready when the field exists.
3. Refuse "funded wallet = authorized" as a compliance stance. The second receipt above is the trap: a conformant-without-judgment implementation auto-pays any invoice from any registered wallet. That's the $78,000 OpenAI Codex failure mode (our breakdown) wearing a standards badge. Gate every conformant payment you emit.
4. Calibrate the decider on your own spend. Today's honest finding is the ceiling of uncalibrated heuristics: 0.35 on everything, fail-closed. Production needs calibrated probabilities per decision — the open-weights JEV-27B family is exactly the input the gate was designed to consume.
5. Engage the Foundation while the window is open. Hursta's mandate is commercialization and the working groups are forming under new leadership — the moment to propose an authorization-layer extension to the standard is now, not after "conformant" already means "auto-pays anything." Join at x402.org and bring the bands: ≥0.80 / 0.50–0.79 / <0.50, with the audit trail attached.

Claim Receipts

Honest caveats

• The appointment facts and Hursta's background come from a single primary source — the Linux Foundation press release — plus wire clones; no independent reporting yet on the Foundation's actual authorization roadmap or conformance criteria.
• The "40 members at launch" figure is via CoinDesk's Oct 1 reporting; the 50+ figure is the Foundation's own claim in the release.
• Gate receipts are from local-heuristic-v1 (uncalibrated, calibrated=false, typesafe_wired=false) — an instrument demo, not a production decider. The identical 0.35 on both questions is disclosed, not hidden.
• The FTC probe dates: announced September 30, 2026 per Reuters (announced via Reuters reporting; covered Oct 1–2). The six-bank audit demand is dated September 22 per prior coverage.
• x402.org is the Foundation's membership site per the Linux Foundation's LinkedIn announcement.

Related: Decision-gated machine payments · Provider Trust · Should AI agents authorize payments? · The $78,000 OpenAI Codex incident · Cloudflare Monetization Gateway · XRPL x402 10 million payments · JEV-27B open decision model

Published Oct 3, 2026 by ScriptMasterLabs — scriptmasterlabs.com