{
  "articles": [
    {
      "author": null,
      "id": "fintech-zen-2026-10-09",
      "markdown_url": "https://qstve.com/corpus/native-fintech-zen/articles/google-s-bid-for-spirit-airlines-working-memory.md",
      "publication": "Fin:Tech & Zen",
      "publication_id": "fintech-zen",
      "published_at": "2026-10-10T19:28:07.787793+00:00",
      "summary": null,
      "title": "Google’s bid for Spirit Airlines’ working memory",
      "url": "https://qstve.com/writing/fintech-zen/google-s-bid-for-spirit-airlines-working-memory/"
    },
    {
      "author": "Dave Glaser",
      "id": "ai-operator-2026-10-09",
      "markdown_url": "https://qstve.com/corpus/native-ai-operator/articles/when-an-ai-task-times-out-check-before-you-retry.md",
      "publication": "AI Operator",
      "publication_id": "ai-operator",
      "published_at": "2026-10-09T18:46:13.297799+00:00",
      "summary": "A missing reply can leave a payment completed and an agent ready to send it again.",
      "title": "When an AI Task Times Out, Check Before You Retry",
      "url": "https://qstve.com/writing/ai-operator/when-an-ai-task-times-out-check-before-you-retry/"
    },
    {
      "author": "Dave Glaser",
      "id": "man-and-machine-2026-10-07",
      "markdown_url": "https://qstve.com/corpus/native-man-and-machine/articles/leave-the-questions-on-the-page.md",
      "publication": "Man and Machine",
      "publication_id": "man-and-machine",
      "published_at": "2026-10-08T14:27:53+00:00",
      "summary": "I asked Oscar to turn our golf schedule into a calendar I could print. One row was cut off, some dates didn’t agree, and one start time looked wrong.",
      "title": "Leave the Questions on the Page",
      "url": "https://qstve.com/writing/man-and-machine/leave-the-questions-on-the-page/"
    },
    {
      "author": "Dave Glaser",
      "id": "the-next-experiment-has-changed-2026-10-04",
      "markdown_url": "https://qstve.com/corpus/native-which-day/articles/the-next-experiment-has-changed.md",
      "publication": "Which Day Are We On?",
      "publication_id": "which-day",
      "published_at": "2026-10-04T09:42:02Z",
      "summary": null,
      "title": "The Next Experiment Has Changed",
      "url": "https://qstve.com/writing/which-day/the-next-experiment-has-changed/"
    },
    {
      "author": null,
      "id": "fintech-zen-2026-10-02",
      "markdown_url": "https://qstve.com/corpus/native-fintech-zen/articles/openai-s-dots-bring-persistent-ai-into-everyday-work-with-accountability-still-to-be-demonstrated.md",
      "publication": "Fin:Tech & Zen",
      "publication_id": "fintech-zen",
      "published_at": "2026-10-03T21:20:11+00:00",
      "summary": null,
      "title": "OpenAI's Dots bring persistent AI into everyday work, with accountability still to be demonstrated",
      "url": "https://qstve.com/writing/fintech-zen/openai-s-dots-bring-persistent-ai-into-everyday-work-with-accountability-still-to-be-demonstrated/"
    },
    {
      "author": "Dave Glaser",
      "id": "ai-operator-2026-10-02",
      "markdown_url": "https://qstve.com/corpus/native-ai-operator/articles/ai-work-needs-proof-not-just-a-good-explanation.md",
      "publication": "AI Operator",
      "publication_id": "ai-operator",
      "published_at": "2026-10-02T13:31:44+00:00",
      "summary": "If you cannot do the work yourself, you still need a clear way to check that it was done right.",
      "title": "AI Work Needs Proof, Not Just a Good Explanation",
      "url": "https://qstve.com/writing/ai-operator/ai-work-needs-proof-not-just-a-good-explanation/"
    },
    {
      "author": "Dave Glaser",
      "id": "man-and-machine-2026-09-30",
      "markdown_url": "https://qstve.com/corpus/native-man-and-machine/articles/the-most-expensive-token-is-the-one-that-did-no-useful-work.md",
      "publication": "Man and Machine",
      "publication_id": "man-and-machine",
      "published_at": "2026-09-30T06:45:00-04:00",
      "summary": "A year of personal AI produced a simple buying rule: match the route to the work, and use no model when none is needed.",
      "title": "The Most Expensive Token Is the One That Did No Useful Work",
      "url": "https://qstve.com/writing/man-and-machine/the-most-expensive-token-is-the-one-that-did-no-useful-work/"
    },
    {
      "author": null,
      "id": "fintech-zen-2026-09-25-style-revision",
      "markdown_url": "https://qstve.com/corpus/native-fintech-zen/articles/when-an-ai-agent-treats-a-security-barrier-as-part-of-the-assignment.md",
      "publication": "Fin:Tech & Zen",
      "publication_id": "fintech-zen",
      "published_at": "2026-09-28T11:16:35Z",
      "summary": null,
      "title": "When an AI agent treats a security barrier as part of the assignment",
      "url": "https://qstve.com/writing/fintech-zen/when-an-ai-agent-treats-a-security-barrier-as-part-of-the-assignment/"
    },
    {
      "author": "Dave Glaser",
      "id": "linkedin-7496551775399768064",
      "markdown_url": "https://qstve.com/corpus/ai-operator/articles/your-agents-prompt-is-not-its-mandate.md",
      "publication": "AI Operator",
      "publication_id": "ai-operator",
      "published_at": "2026-08-21",
      "summary": "Once software can trade or pay, financial authority has to survive outside the conversation.",
      "title": "Your Agent's Prompt Is Not Its Mandate",
      "url": "https://qstve.com/writing/linkedin/ai-operator/your-agents-prompt-is-not-its-mandate/"
    },
    {
      "author": "Dave Glaser",
      "id": "linkedin-7494015042854703104",
      "markdown_url": "https://qstve.com/corpus/ai-operator/articles/your-ai-agent-is-missing-the-relationships.md",
      "publication": "AI Operator",
      "publication_id": "ai-operator",
      "published_at": "2026-08-14",
      "summary": "Why the next practical control layer is a small map connecting intent, authority, action, and evidence. We call these Graphs.",
      "title": "Your AI Agent Has the Data. It Is Missing the Relationships. Let's talk about Graphs.",
      "url": "https://qstve.com/writing/linkedin/ai-operator/your-ai-agent-is-missing-the-relationships/"
    },
    {
      "author": "Dave Glaser",
      "id": "linkedin-7491478399400583168",
      "markdown_url": "https://qstve.com/corpus/ai-operator/articles/ai-agents-can-stay-on-the-case.md",
      "publication": "AI Operator",
      "publication_id": "ai-operator",
      "published_at": "2026-08-07",
      "summary": "Long-running financial agents need authority that expires, narrows, and can be revoked without dismantling the workflow.",
      "title": "AI Agents Can Stay on the Case. Their Financial Permission Shouldn't.",
      "url": "https://qstve.com/writing/linkedin/ai-operator/ai-agents-can-stay-on-the-case/"
    }
  ],
  "canonical_url": "https://qstve.com/reading/",
  "field_guides": [
    {
      "guidance_note": "The checklist and prompt are editorial adaptations, not tested transcripts. Last week's result-checking principle is applied here to lost payment replies.",
      "id": "2026-10-09-when-an-ai-task-times-out-check-before-you-retry",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-10-09-when-an-ai-task-times-out-check-before-you-retry.pdf",
      "published_at": "2026-10-09",
      "summary": "A payment can complete before its reply reaches an agent. Keep the original payment identity across restarts. Use the provider's supported protected retry, or pause and assign a person to check an unknown result before allowing a fresh payment.",
      "title": "When an AI Task Times Out, Check Before You Retry",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-10-09-when-an-ai-task-times-out-check-before-you-retry"
    },
    {
      "guidance_note": "Adapted from the approved October 2 article and retained companion. The companion originally carried October 9; this guide uses the published edition date. Prompts are templates, not tested transcripts or execution authority.",
      "id": "2026-10-02-check-the-result-and-keep-the-proof",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-10-02-check-the-result-and-keep-the-proof.pdf",
      "published_at": "2026-10-02",
      "summary": "Choose the proof before an agent begins. Verify the changed result, follow evidence for important claims and measure the time spent reviewing and repairing the work.",
      "title": "AI Work Needs Proof, Not Just a Good Explanation",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-10-02-check-the-result-and-keep-the-proof"
    },
    {
      "guidance_note": "Adapted design guidance from the published September 25 edition. It does not establish legal liability or certify a payment service.",
      "id": "2026-09-25-govern-the-mandate-before-the-payment",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-09-25-govern-the-mandate-before-the-payment.pdf",
      "published_at": "2026-09-25",
      "summary": "When an agent can select a payment method or establish recurring authority, govern the mandate before a transaction exists. Preserve that authority through execution and reconciliation.",
      "title": "The New Payment Control Point",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-09-25-govern-the-mandate-before-the-payment"
    },
    {
      "guidance_note": "The contract fields, failure drills and rollout sequence are practical operator guidance derived from the article’s Operator’s Check and Money Move.",
      "id": "2026-09-18-bind-mandate-to-outcome",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-09-18-bind-mandate-to-outcome.pdf",
      "published_at": "2026-09-18",
      "summary": "Know-Your-Agent controls can identify and constrain the software actor. The enterprise still needs one continuous record connecting the principal, mandate, policy decision, settlement, delivered result, exception owner and recovery path.",
      "title": "The Agent Has an Identity. The Company Still Owns the Decision",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-09-18-bind-mandate-to-outcome"
    },
    {
      "guidance_note": "The bounded cycle, independent evaluation and canary rollback are operational guidance. They do not establish that recursive self-improvement is production-safe.",
      "id": "2026-09-11-measure-the-handoff-not-the-model",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-09-11-measure-the-handoff-not-the-model.pdf",
      "published_at": "2026-09-11",
      "summary": "Agents can study outcomes and propose better procedures. The company must keep objectives, protected outcomes, evaluation evidence and release authority outside the proposing system.",
      "title": "When AI Starts Rewriting the Company",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-09-11-measure-the-handoff-not-the-model"
    },
    {
      "guidance_note": "The action-envelope schema and stale-field test are operational guidance; reasoning traces remain one sensor among several.",
      "id": "2026-09-04-make-exceptions-expire",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-09-04-make-exceptions-expire.pdf",
      "published_at": "2026-09-04",
      "summary": "A model can behave better overall while becoming harder to inspect in the exceptional case. Reasoning traces remain useful sensors; durable authority comes from an external action envelope checked at execution.",
      "title": "When the Reasoning Trace Goes Quiet",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-09-04-make-exceptions-expire"
    },
    {
      "guidance_note": "The duplicate-environment exercise and boundary design are defensive operational guidance, not instructions for live intrusion testing.",
      "id": "2026-08-28-turn-oversight-into-a-shift",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-08-28-turn-oversight-into-a-shift.pdf",
      "published_at": "2026-08-28",
      "summary": "An agent may accept a persuasive description of its environment. A live execution boundary must obtain environment identity, approval, limits and revocation from trusted state outside the conversation.",
      "title": "“This Is a Test” Is Not an Authorization",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-08-28-turn-oversight-into-a-shift"
    },
    {
      "guidance_note": "The seven-line mandate and negative tests are operational guidance. Provider controls still depend on configuration and enforcement.",
      "id": "2026-08-21-rehearse-the-handoff",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-08-21-rehearse-the-handoff.pdf",
      "published_at": "2026-08-21",
      "summary": "A detailed prompt remains conversational context. Consequential financial authority needs a separate, machine-checkable mandate evaluated at the tool boundary for every proposed action.",
      "title": "Your Agent's Prompt Is Not Its Mandate",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-08-21-rehearse-the-handoff"
    },
    {
      "guidance_note": "The small-map exercise is operational guidance; it does not prescribe a graph database or treat inferred edges as approved facts.",
      "id": "2026-08-14-put-expert-judgment-on-the-record",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-08-14-put-expert-judgment-on-the-record.pdf",
      "published_at": "2026-08-14",
      "summary": "Records distributed across finance, identity, workflow and approval systems cannot answer an authority question until their relationships are explicit. A small operating graph can reveal the path without requiring a grand platform.",
      "title": "Your AI Agent Has the Data. It Is Missing the Relationships.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-08-14-put-expert-judgment-on-the-record"
    },
    {
      "guidance_note": "The lease model, event triggers and pause test are operational guidance derived from the article.",
      "id": "2026-08-07-practice-on-the-replica",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-08-07-practice-on-the-replica.pdf",
      "published_at": "2026-08-07",
      "summary": "Long-running workflows can retain useful context while financial authority expires. Treating permission as a lease keeps a persistent agent from acting after balances, policy, risk or customer intent have changed.",
      "title": "AI Agents Can Stay on the Case. Their Financial Permission Shouldn't.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-08-07-practice-on-the-replica"
    },
    {
      "guidance_note": "The five-part packet and reversal drill are operator guidance from the issue's Money Move.",
      "id": "2026-07-31-automate-the-briefing-not-the-verdict",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-07-31-automate-the-briefing-not-the-verdict.pdf",
      "published_at": "2026-07-31",
      "summary": "Agent identity and tokenised checkout make software legible at the front door. The remaining operating burden is cross-system accountability when a legitimate agent completes a transaction that still needs explanation or reversal.",
      "title": "AI Agents Can Checkout. Nobody Owns the Exception Packet.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-07-31-automate-the-briefing-not-the-verdict"
    },
    {
      "guidance_note": "The containment receipt, threat model and shutdown rehearsal are operational guidance; they do not assert that deployed agents routinely evade controls.",
      "id": "2026-07-24-work-backward-from-go",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-07-24-work-backward-from-go.pdf",
      "published_at": "2026-07-24",
      "summary": "A reported evaluation incident showed why declared permission is weaker than enforced containment. Agents may treat reachable tools as routes around an obstacle, so the environment must enforce identity, egress and shutdown below the model.",
      "title": "The Agent Was Denied Internet. It Found A Door Anyway.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-07-24-work-backward-from-go"
    },
    {
      "guidance_note": "The protocol-neutral ledger and cross-adapter revocation test are operational guidance based on that distinction.",
      "id": "2026-07-17-show-your-system-where-it-is-wrong",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-07-17-show-your-system-where-it-is-wrong.pdf",
      "published_at": "2026-07-17",
      "summary": "Payment protocols can carry machine transactions, yet protocol capability does not establish business authority. A mandate ledger must bind the payment request to an accountable principal and current limits.",
      "title": "The Protocol Can Pay. Finance Still Needs The Mandate.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-07-17-show-your-system-where-it-is-wrong"
    },
    {
      "guidance_note": "Freshness windows, snapshots and transformation records are implementation guidance drawn from the source-receipt framework.",
      "id": "2026-07-10-proof-on-the-actual-route",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-07-10-proof-on-the-actual-route.pdf",
      "published_at": "2026-07-10",
      "summary": "An authorised agent can still act on stale catalogues, policies, quotes or public records. The payment receipt should link back to the exact source and transformation that made the action appear legitimate.",
      "title": "The Agent Read The Record. Finance Needs The Receipt.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-07-10-proof-on-the-actual-route"
    },
    {
      "guidance_note": "Mandate fields and source snapshots are operational guidance for making delegated commerce reviewable.",
      "id": "2026-07-03-turn-ai-work-into-a-receipt",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-07-03-turn-ai-work-into-a-receipt.pdf",
      "published_at": "2026-07-03",
      "summary": "Shopping agents are entering merchant discovery and live transactions. When the conversation becomes the storefront, finance must preserve who the agent represented, the permission it carried and the evidence behind the purchase.",
      "title": "The Agent Is Now The Storefront. Finance Still Needs The Mandate.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-07-03-turn-ai-work-into-a-receipt"
    },
    {
      "guidance_note": "The exception packet, routing approach and checklist are practical operating guidance based on the issue's Money Move.",
      "id": "2026-06-26-run-the-rollback-drill",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-06-26-run-the-rollback-drill.pdf",
      "published_at": "2026-06-26",
      "summary": "Agent payments now span cards, accounts, wallets, stablecoins and treasury APIs. Fast authorisation still leaves finance with a human-speed problem when business context is stale or a technically valid payment is wrong.",
      "title": "AI Payments Reached Machine Speed. The Exception Desk Did Not.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-06-26-run-the-rollback-drill"
    },
    {
      "guidance_note": "Ledger fields and run-close practices are operator guidance distilled from the article's delegation argument.",
      "id": "2026-06-19-assignment-ledger",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-06-19-assignment-ledger.pdf",
      "published_at": "2026-06-19",
      "summary": "Agents are taking multi-step work across documents, data and finance. Recurring work needs a durable delegation record so another operator can see the assignment, authority, evidence, changes and recovery route.",
      "title": "AI Coworkers Got Assignments. Nobody Built The Assignment Ledger.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-06-19-assignment-ledger"
    },
    {
      "guidance_note": "The schema, replay test and append-only implementation hints are operational guidance, not claims about a named provider's controls.",
      "id": "2026-06-12-finance-needs-the-event-log",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-06-12-finance-needs-the-event-log.pdf",
      "published_at": "2026-06-12",
      "summary": "As commerce enters cameras, assistants and programmable settlement, a conventional receipt cannot explain machine judgement. Finance needs an event log that preserves intent, authority, evidence, decision and outcome.",
      "title": "The Receipt Is Becoming the Interface. Finance Needs the Event Log.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-06-12-finance-needs-the-event-log"
    },
    {
      "guidance_note": "The revocation-card pattern and checklist are operational guidance derived from the issue's Money Move and Operator's Log.",
      "id": "2026-06-05-finance-needs-the-off-switch",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-06-05-finance-needs-the-off-switch.pdf",
      "published_at": "2026-06-05",
      "summary": "Payment agents are reaching credentials, wallets and settlement systems. The durable control is a narrow mandate with retained evidence, human thresholds and a revocation route that works before the next transaction.",
      "title": "The Agent Is Becoming the Approver. Finance Needs the Off Switch.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-06-05-finance-needs-the-off-switch"
    },
    {
      "guidance_note": "Derived operational guidance. Provider examples are historical context; confirm present capabilities before selecting an implementation.",
      "id": "2026-05-31-keep-the-approval-path-with-the-payment",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-05-31-keep-the-approval-path-with-the-payment.pdf",
      "published_at": "2026-05-31",
      "summary": "A payment receipt records movement of money. A delegated-spend record should also explain the purpose, policy, permission and evidence behind the action.",
      "title": "AI Agents Can Pay. The Ledger Still Cannot Explain Why.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-05-31-keep-the-approval-path-with-the-payment"
    },
    {
      "guidance_note": "This guide adapts the live-state comparison method. Historical provider claims and counts describe that incident, not current service behavior.",
      "id": "2026-04-10-compare-live-state-with-approved-state",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-04-10-compare-live-state-with-approved-state.pdf",
      "published_at": "2026-04-10",
      "summary": "Configuration records can drift from running agents. Compare the actual jobs, model and access settings with the approved configuration, and retain the differences.",
      "title": "Nine Crons Were Running. I Thought It Was Four.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-04-10-compare-live-state-with-approved-state"
    },
    {
      "guidance_note": "This is an adaptation of its operating checklist. Passing a checklist alone does not authorize unattended production use.",
      "id": "2026-04-03-build-the-operating-layer-with-the-agent",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-04-03-build-the-operating-layer-with-the-agent.pdf",
      "published_at": "2026-04-03",
      "summary": "An agent needs a usable success record and recoverable failure state as part of its design. Build those alongside the task, then test the whole run.",
      "title": "AI Agents Got Into Finance. Nobody Built the Operating Layer.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-04-03-build-the-operating-layer-with-the-agent"
    },
    {
      "guidance_note": "Derived design guidance, not a claim that all fraud products lack agent controls or a determination of dispute rights.",
      "id": "2026-03-27-make-agent-spending-scope-enforceable",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-03-27-make-agent-spending-scope-enforceable.pdf",
      "published_at": "2026-03-27",
      "summary": "A transaction can look acceptable to a fraud system while exceeding the instruction given to an agent. Make delegated spending authority explicit and enforceable.",
      "title": "AI Agents Got Credit Cards. Nobody Built the Fraud Stack.",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-03-27-make-agent-spending-scope-enforceable"
    },
    {
      "guidance_note": "This checklist adapts the governance questions; it does not establish current protocol availability or legal liability.",
      "id": "2026-03-20-separate-task-and-payment-permission",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-03-20-separate-task-and-payment-permission.pdf",
      "published_at": "2026-03-20",
      "summary": "Giving an agent a task does not settle which purchases it may make. Define spending permission and exception ownership before connecting payment tools.",
      "title": "The Week Machines Got Their Own Money",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-03-20-separate-task-and-payment-permission"
    },
    {
      "guidance_note": "Derived operating guidance. Confirm current rail-specific timing, return rights and recovery rules; the headline is not a universal service limit.",
      "id": "2026-03-13-measure-risk-at-payment-speed",
      "pdf_url": "https://qstve.com/writing/linkedin/ai-operator/field-guides/2026-03-13-measure-risk-at-payment-speed.pdf",
      "published_at": "2026-03-13",
      "summary": "Fast settlement makes the timing and quality of fraud decisions an operating concern. Measure decision latency, evidence freshness and the cost of wrongly blocked payments together.",
      "title": "The 400-Millisecond Problem: Why Real-Time Payments Broke Bank Fraud Models",
      "url": "https://qstve.com/writing/ai-operator/field-guides/#2026-03-13-measure-risk-at-payment-speed"
    }
  ],
  "schema_version": 1,
  "scope": "Accepted complete QSTVE essays and accepted AI Operator Field Guides; external archive references are not complete hosted text.",
  "title": "QSTVE public reading catalog",
  "topics": [
    {
      "article_ids": [
        "ai-operator-2026-10-02",
        "linkedin-7496551775399768064",
        "linkedin-7494015042854703104",
        "linkedin-7491478399400583168"
      ],
      "guide_ids": [
        "2026-10-02-check-the-result-and-keep-the-proof",
        "2026-08-21-rehearse-the-handoff",
        "2026-08-14-put-expert-judgment-on-the-record"
      ],
      "question": "What proves that an AI agent did the right work?",
      "related": [
        {
          "reason": "Measure whether a better tool produces better work.",
          "slug": "ai-changing-work"
        },
        {
          "reason": "Follow authority through a consequential financial action.",
          "slug": "payments-infrastructure"
        }
      ],
      "sections": [
        {
          "heading": "Start with the result",
          "sources": [
            "ai-operator-2026-10-02",
            "2026-10-02-check-the-result-and-keep-the-proof"
          ],
          "text": "A convincing account of activity is useful, but it cannot replace a changed record, a delivered artifact or a checked claim. QSTVE's October 2 essay argues that review and repair belong in the cost of automation. Define completion proof before the agent begins, then check the destination rather than its progress report."
        },
        {
          "heading": "Keep permission outside the conversation",
          "sources": [
            "linkedin-7496551775399768064",
            "linkedin-7491478399400583168"
          ],
          "text": "A prompt describes the work. A mandate defines the authority to act. The published AI Operator essays separate these roles: financial permission needs current limits, expiry and revocation at the action boundary, even when an agent remains active for days."
        },
        {
          "heading": "Connect the evidence",
          "sources": [
            "linkedin-7494015042854703104",
            "2026-08-14-put-expert-judgment-on-the-record"
          ],
          "text": "A small operating graph can link intent, authority, action and evidence across records that otherwise stand alone. The purpose is practical: let an operator follow the path from an assignment to the permission and outcome that justify it. Start with one workflow and its consequential decisions."
        }
      ],
      "slug": "ai-agent-accountability",
      "summary": "Choose the evidence before the task begins. Connect the agent's assignment to its authority, the action it took and the result another person can check.",
      "title": "AI agent accountability",
      "url": "https://qstve.com/topics/ai-agent-accountability/"
    },
    {
      "article_ids": [
        "man-and-machine-2026-09-30",
        "the-next-experiment-has-changed-2026-10-04",
        "ai-operator-2026-10-02"
      ],
      "guide_ids": [
        "2026-09-11-measure-the-handoff-not-the-model",
        "2026-10-02-check-the-result-and-keep-the-proof",
        "2026-04-10-compare-live-state-with-approved-state"
      ],
      "question": "How do we tell whether AI is making work better?",
      "related": [
        {
          "reason": "Choose completion proof and an independent check.",
          "slug": "ai-agent-accountability"
        },
        {
          "reason": "Explore work where permission and recovery matter alongside speed.",
          "slug": "payments-infrastructure"
        }
      ],
      "sections": [
        {
          "heading": "Choose the route after the job",
          "sources": [
            "man-and-machine-2026-09-30"
          ],
          "text": "In Man and Machine, Dave Glaser describes a deterministic task that still called a premium model. The work finished, but the model added nothing. His resulting rule is to begin with the need for judgement, structure, privacy or speed, then choose the smallest route that can do the work well. Some tasks need no model."
        },
        {
          "heading": "Follow improvements into the next tools",
          "sources": [
            "the-next-experiment-has-changed-2026-10-04"
          ],
          "text": "The Next Experiment Has Changed follows dated reports of AI improving research software and computing infrastructure. Its distinction matters: a faster calculation can change what researchers can explore, but does not establish faster whole research programmes or a dependable exponential growth rate. Developer reports and preprints retain their stated limits."
        },
        {
          "heading": "Close the improvement loop with an outside check",
          "sources": [
            "2026-09-11-measure-the-handoff-not-the-model"
          ],
          "text": "The accepted Field Guide for When AI Starts Rewriting the Company recommends fixing the objective and protected outcomes, freezing a baseline, limiting the change surface and using independent evaluation. A narrow canary and a demonstrated rollback turn a proposal into a bounded test. This is operating guidance, not proof that autonomous improvement is production-safe."
        }
      ],
      "slug": "ai-changing-work",
      "summary": "Begin with the job and a way to check it. Compare useful results, the cost of review and the limits of the evidence before choosing a model or expanding a workflow.",
      "title": "AI changing work",
      "url": "https://qstve.com/topics/ai-changing-work/"
    },
    {
      "article_ids": [
        "linkedin-7496551775399768064",
        "linkedin-7491478399400583168",
        "linkedin-7494015042854703104"
      ],
      "guide_ids": [
        "2026-09-18-bind-mandate-to-outcome",
        "2026-05-31-keep-the-approval-path-with-the-payment",
        "2026-03-13-measure-risk-at-payment-speed",
        "2026-06-26-run-the-rollback-drill"
      ],
      "question": "What has to travel with an agent-initiated payment?",
      "related": [
        {
          "reason": "Connect each action to current authority and proof.",
          "slug": "ai-agent-accountability"
        },
        {
          "reason": "Measure useful work without overlooking review and recovery.",
          "slug": "ai-changing-work"
        }
      ],
      "sections": [
        {
          "heading": "Distinguish access from authority",
          "sources": [
            "linkedin-7496551775399768064",
            "2026-03-20-separate-task-and-payment-permission",
            "2026-03-27-make-agent-spending-scope-enforceable"
          ],
          "text": "An account connection or working payment tool does not establish authority for every purchase. AI Operator's complete essays and accepted guides distinguish task permission from spending permission. Bind the business purpose to permitted counterparties, amounts, time limits and a revocation path that can be checked outside the prompt."
        },
        {
          "heading": "Keep the approval path with the payment",
          "sources": [
            "2026-05-31-keep-the-approval-path-with-the-payment",
            "2026-09-18-bind-mandate-to-outcome",
            "2026-06-26-run-the-rollback-drill"
          ],
          "text": "The accepted guides recommend retaining intent, policy, permission and evidence with the action. They also connect settlement to the delivered business result. A technically valid payment can still require explanation or recovery; assign the exception owner and test reconciliation before expanding an agent's scope."
        },
        {
          "heading": "Measure the whole decision path",
          "sources": [
            "2026-03-13-measure-risk-at-payment-speed"
          ],
          "text": "The real-time payments Field Guide recommends measuring decision latency, data freshness, false positives and customer friction together. Its headline is not a universal payment service limit. Timing, return rights and recovery procedures must be confirmed for the particular rail and use case."
        }
      ],
      "slug": "payments-infrastructure",
      "summary": "A payment records money movement. Accountable delegated spending also needs a principal, a current mandate, evidence for the decision and an owner for exceptions and recovery.",
      "title": "Payments infrastructure",
      "url": "https://qstve.com/topics/payments-infrastructure/"
    }
  ]
}