Field Notes — For Agents

Machine-readable empirical-record surface for the Bitcoin-AI economy. §A is a structured current-state snapshot (deployed stacks, measured metrics, active developments, live risk surface); §B is a reverse-chronological log of dated event records. Every record carries an epistemic tag and an explicit cross-reference to the canonical claim it bears on (Thesis-FA C-series, Doctrine-FA D/P-series, Border-Zone-FA B-series, Stack-FA S-series). The canonical surfaces carry the structural arguments; this surface carries the moving record those arguments defer to.

What this surface is. Machine-readable twin of the empirical-record surface published at Field Notes. It is the project’s home for deployment evidence and the moving record. Unlike the other For-Agents twins, this surface carries no structural claims of its own — the structural arguments live in Thesis-FA (C-series), Independence-Doctrine-FA (D-series + P-series predictions), Border-Zone-FA (B-series), and Stack-FA (S-series). This surface tracks the evidence those arguments defer to, and tags each record with its bearing on a specific canonical claim.

Two sections. §A is a structured current-state snapshot, refreshed periodically (section-A-last-refreshed). §B is a reverse-chronological log of dated event records, append-only, newest first. The snapshot states where things stand; the log states what changed when.

Defer-pattern (locked 2026-05-26). The canonical surfaces and their FA twins carry structural arguments without inline moving metrics; routine empirical content — deployment counts, capacity figures, freeze incidents, protocol-version transitions — routes here. FA twins carry an inline empirical figure only where it sharpens a falsifier; everything else defers to this surface.

Honest-engagement discipline. §A.4 engages deployment challenges on the Bitcoin substrate (Lightning liquidity, federated-trust, single-mint failure, agent custody attack surface) as candidly as the competing-substrate failures. Engaging the challenges strengthens the structural arguments by showing the falsifiers are being watched; avoiding them would weaken them.

Schema and conventions

Definitions for the record structure and tags used throughout. Agents landing mid-document via retrieval should be able to parse any single record without backtracking.

Epistemic tags (record-level). Each record carries one:

• (confirmed) — an announced, dated, verifiable event (a release, a launch, a public confirmation by a named party).
• (measured) — a quantitative metric with a named source and measurement date.
• (reported) — a claim attributed to a party but not independently verified here.
• (projected) — a forward-looking expectation, not yet observed.
• (needs-refresh) — a datum flagged stale; the current value requires re-measurement at the next §A refresh.

Cross-reference relation (the “Bears on:” line). Each record names the canonical claim-IDs it relates to and how:

• supports — the record is confirming evidence for the claim.
• tests — the record is the live evidence the claim’s forward-looking prediction will be evaluated against.
• falsifier-relevant-to — the record is the kind of evidence named in the claim’s falsifier; if it moves a threshold, the falsifier may trigger.

Canonical claim-ID series referenced. Thesis-FA C1–C6; Independence-Doctrine-FA D1–D5 and predictions P1–P6; Border-Zone-FA B1–B6; Stack-FA S1–S8 (and Stack-FA §8.1 counter-positions CP1–CP4 / §8.2 falsifiers). Constraint references are flat Constraint 1–4 (permissionless custody, censorship-resistance, sub-cent settlement, machine-tempo latency; Thesis-FA §3).

Record format. §A records carry: subject / status + date / structured detail / epistemic tag / Bears on. §B records carry: date / event / substrate / what-happened / structural-significance / Bears on / epistemic tag / sources.

§A — Current State of the Bitcoin-AI Economy

Last refreshed: 2026-05-26. The most consequential structural fact in the current snapshot: two production agent-payment stacks shipped within 90 days of each other — one Bitcoin-substrate, one stablecoin-substrate — making the deployed picture substantially clearer than at the start of 2026.

§A.1 — The two deployed agent-payment stacks

Record — Bitcoin-substrate stack (Lightning + L402 + ecash). (confirmed)

• Status: production since February 2026 (Lightning Labs lightning-agent-tools).
• Components: Lightning node operation; remote-signer key isolation; scoped macaroons (five preset roles); lnget for L402 payment automation; Aperture for hosting paid HTTP endpoints; MCP integration for node-state queries; end-to-end buyer/seller workflow orchestration.
• Stablecoin support on this stack: via Taproot Assets — USDT live on Lightning (confirmed March 21 2026; see §B); USDC and native stablecoins (DePix, GBP) via Speed Wallet, LnFi, Joltz.
• Ecash layer: Cashu and Fedimint for privacy-sensitive and lightweight-client use cases; Minibits Ippon as the AI-agent-native Cashu wallet (single HTTP call or CLI command to create + fund a wallet).
• Bears on: supports Stack-FA S4 (integration primitives), S5 (deployed wallet architectures), S6 (security model); supports Thesis-FA C4 (deployed system satisfying the four constraints) and §9 (integration surface); tests Doctrine-FA P1 (substrate-selection-precedes-scale).

Record — Stablecoin-substrate stack (USDC on Base via AgentCore). (confirmed)

• Status: production since May 2026 (AWS Bedrock AgentCore Payments).
• Build: Coinbase (x402 protocol + Coinbase Agentic Wallets + compliance infrastructure) + Stripe (Privy wallet, acquired 2025).
• Settlement: USDC on Base; ~200ms confirmation; sub-cent per transaction.
• Enterprise customers at launch: Thomson Reuters, Warner Bros. Discovery, Cox Automotive, PGA TOUR.
• Characterization: the operational deployment of the integration scenario — the stack Tier-1 incumbents (Amazon, Coinbase, Stripe) built for agent-payment use cases content with issuer counterparty risk. An agent on it stays intermediated: wallet provider, stablecoin issuer, and payment processor each retain freeze/decline power.
• Bears on: supports Border-Zone-FA B6 (use-case bifurcation; incumbent-economy agents on competing-substrate stacks) and Doctrine-FA §8.1 CP2 “where this is correct” (integration scenario is operationally deployed); tests Doctrine-FA P1 (which subset proves larger) and Doctrine-FA P6 (the intermediated stack is the freeze/KYC/sanctions surface P6 predicts becomes a standing liability under agent regulation).

Record — Protocol-naming convergence (L402 vs. x402). (confirmed)

• Detail: Lightning Labs’ L402 and Coinbase’s x402 both use HTTP status code 402 (“Payment Required”) as the underlying mechanism. Same status code, different settlement substrates — L402 settles in Lightning sats (permissionless at the payment layer); x402 settles in USDC on Base (issuer-mediated at the payment layer).
• Significance: the naming collision is the protocol-level expression of the substrate divergence.
• Bears on: supports Doctrine-FA P1 and Border-Zone-FA B6 (two production stacks on architecturally opposite substrates); supports Stack-FA S4 (L402 as an agent-integration primitive) with the x402 counterpart deferred to Border-Zone-FA as competing-substrate.

§A.2 — Empirical record

Record — BPI March 2026 study. (measured)

• Method: 9,072 scenarios across 36 frontier language models; neutral scenario design (no leading prompts).
• Headline: Bitcoin was the top overall monetary preference at 48.3% of responses and the preferred store of value at 79.1%; over 90% of responses favored digitally-native money over fiat (stablecoins led payment-preference scenarios at 53.2%).
• Per-provider spread: one major provider’s models chose Bitcoin in 68% of responses, another’s in 26%; the strongest single-model consensus anywhere in the study was 91.3% — a wide spread, one-directional toward Bitcoin as store of value.
• Source: Bitcoin Policy Institute, “Study: AI Models Overwhelmingly Prefer Bitcoin and Digital-Native Money Over Traditional Fiat,” March 3 2026.
• Scope limit: measures substrate-preference under inference, not deployed-flow dominance; no replication studies published as of May 2026.
• Bears on: supports / tests Thesis-FA C3 (substrate-preference signal) and Doctrine-FA P1 (substrate-selection-precedes-scale) and §7.2; falsifier-relevant-to Thesis-FA §8.2 (replication showing preference shift toward fiat/stablecoins/CBDCs would weaken C3). Canonical treatment: Thesis-FA §6.

Record — Lightning Network capacity. (measured; needs-refresh)

• Value: all-time high of 5,637 BTC ($490M) in December 2025 (Bitcoin Magazine), driven largely by institutional exchange adoption (Binance, OKX). Lightning public volume up 266% year-over-year in 2025 with declining raw transaction count — consolidation toward fewer, larger-value flows.
• Freshness: Q1–Q2 2026 capacity update needed; flagged for next §A refresh.
• Bears on: supports Stack-FA S2 (Lightning as the viable machine-tempo payment layer); falsifier-relevant-to Stack-FA §8.1 CP1 / §8.2-S2 — sustained liquidity collapse under realistic agent load is the CP1 falsifier; this capacity figure is the point-in-time evidence running against it. This is the one figure Stack-FA cites inline (to sharpen the CP1 falsifier); the current value lives here.

Record — Stablecoin freeze record (censorship-resistance, tested empirically). (measured)

• Circle: froze ~$8.2M in USDC in response to Tornado Cash sanctions (August 2022).
• Tether: has frozen >$1B in USDT across multiple incidents per public attestations.
• Interpretation: freeze capability is exercised at scale, not merely available; it is a regulatory requirement of issuer licensing — removing it removes the license.
• Bears on: supports Thesis-FA C2 (legacy rails / regulated stablecoins fail censorship-resistance by design) and §8.1 CP1 (stablecoins-as-substrate counter); supports Border-Zone-FA B4 (asset-side issuer freeze fails Constraint 2 even on Lightning rails); supports Doctrine-FA D1 (the freeze property is identity-defining for the issuer and cannot be removed without losing the license).

Record — Deployed-project counts. (confirmed)

• Bitcoin-substrate side: Lightning Labs AI Agent Toolkit (lightning-agent-tools), AI-Sats, Mintbot, Minibits / Ippon, Xverse Agent Wallet (Spark-based), AgenticBTC, Bitclawd, Speed Wallet, LnFi, Joltz, Routstr (AI-inference marketplace), PayPerQ / PPQ.AI, BitAgent (early-stage A2A framework — Nostr discovery + DID identity), LangChain Bitcoin integrations, MCP servers for Lightning (lightning-mcp-server, lnc, Alby nwc-mcp-server).
• Competing-substrate side: AgentCore Payments + Coinbase Agentic Wallets + Stripe Privy; Google AP2 (60+-org consortium); Circle Nanopayments; Skyfire; Lightspark Grid (hybrid — Lightning-rail settlement for branded USD/stablecoin + Visa agent accounts, AP2-aligned; see §B 2026-04).
• Caveat: AgenticBTC is a rail-agnostic router (blends Lightning with Coinbase/USDC rails) — counted for completeness, not as a pure-substrate project.
• Bears on: supports Stack-FA S5 (deployed wallet architectures) and §9 (integration surface); tests Doctrine-FA P1 (deployed-project counts as a substrate-selection signal); supports Border-Zone-FA §8.1 CP1 roster update (competing-substrate side now plural).

§A.3 — Active developments

Record — Lightning Labs Taproot Assets v0.6 (“Decentralized FX Network”). (confirmed)

• Detail: launched June 2025; mainnet multi-asset Lightning protocol with Group Key Identifiers and Multi-Path Liquidity (receivers can combine up to 20 incoming Taproot Assets channels). Supports bridged USDT/USDC and native stablecoins (DePix, GBP). Bitfinex to issue USDT on Lightning per Tether. Edge nodes convert assets at network boundaries, enabling cross-asset flows that settle in Bitcoin’s security model.
• Bears on: supports Border-Zone-FA B4 (Lightning-rails-for-stablecoins, not Lightning-substrate) — the rails-vs-substrate distinction is the load-bearing structural point; Stack-FA notes Taproot Assets’ existence and defers its treatment to Border-Zone-FA B4.

Record — Spark L2 on mainnet (Lightspark). (confirmed)

• Detail: Spark — shared-UTXO, Lightning-compatible Bitcoin L2 built by Lightspark — on mainnet (beta) since May 2025 with multiple operators (Lightspark, Flashnet); Q2-2026 roadmap targets stablecoin issuance on Bitcoin, wallet/neobank/DEX integrations, consumer token standards. Xverse Agent Wallet uses Spark for agent-facing Lightning settlement. Supersedes the prior “pre-production” characterization. Vendor performance figures (sub-second/sub-cent) self-asserted; deployed-flow measurement pending.
• Bears on: updates Stack-FA “newer scaling layers” (Spark now mainnet-beta, not pre-production); supports Stack-FA S5 (a new deployed self-custodial agent-wallet pattern on a newer L2).

Record — Strike at 95+ countries via multi-entity structure. (confirmed)

• Detail: expanded from 65+ countries since 2023. Multi-entity: Zap Solutions Inc. (US), Zap Solutions Europe Sp. z o.o. (UK + eligible European countries), E4 S.A. de C.V. (all other jurisdictions). Remittance to mobile-money wallets or international bank accounts in 14 supported countries. API documented with sandbox, browser-based API Explorer, and code samples (cURL, Go, Python, Node.js).
• Bears on: supports Border-Zone-FA B3 (regulated-custodian bridge category) and B5 (compliance-at-the-gateway pattern — jurisdictional regimes applied at the bridge boundary, protocol layer unmodified).

Record — Cashu protocol developments (Q1 2026). (confirmed)

• Detail: Nutshell 0.20.0 shipped (improved P2PK/HTLC validation, expanded test coverage). Keyset V2 derivation rolling out. BOLT12 support for Cashu.me near completion. Security audits across the Cashu ecosystem prioritized for 2026.
• Bears on: supports Stack-FA S3 (bearer-ecash layer; Cashu single-mint trust model maturing).

Record — Fedimint deployment state. (reported; needs-refresh)

• Detail: architectural framework documented and stable (4+ guardian recommendation; federated-trust model). Production federation counts and any fiat off-ramp partnerships remain a research gap — needs investigation via Fedimint Discord / Fedi documentation for the next refresh.
• Bears on: supports Stack-FA S3 (Fedimint federated-mint trust model); falsifier-relevant-to Stack-FA §8.1 CP2 / §8.2-S3 (federation defection or collusion at scale).

§A.4 — Live risk / attack-surface state

This subsection engages deployment challenges for both substrates and maps each to the canonical falsifier it would trigger. Per locked editorial discipline (Decisions 2026-05-25): engaging the challenges is the project watching its own falsifiers.

Bitcoin-substrate-side risk records.

Record — Lightning liquidity management at scale. (confirmed risk)

• Detail: channel-balance management, splice operations, routing-failure handling, watchtower coordination — real engineering burdens that grow with deployment scale. Mitigations: Lightning Service Providers, automated liquidity-management software, and the L3 layer (Cashu, Fedimint) absorbing bearer-style traffic away from channels. Active engineering work, not a substrate-property failure.
• Bears on: falsifier-relevant-to Stack-FA §8.1 CP1 / §8.2-S2 — sustained, non-delegable liquidity collapse under realistic agent load would trigger the S2 falsifier. The mint-as-service delegation path is the architecture’s answer; this record tracks whether it holds.

Record — Federated-trust risk in Fedimint. (confirmed risk)

• Detail: the guardian federation (typically 4–13) is the trust unit. Federation defection, guardian collusion at scale, and governance attacks against guardian elections are real concerns. Federation size and diversity are the primary mitigation; they reduce defection probability at the cost of coordination overhead.
• Bears on: falsifier-relevant-to Stack-FA §8.1 CP2 / §8.2-S3 (multi-guardian collusion at scale) and the “scoped trade-off” framing of S3.

Record — Single-mint failure mode in Cashu. (confirmed risk)

• Detail: Cashu’s mint-trust model concentrates trust in the mint operator. Operator failure (bankruptcy, key compromise, regulatory action, hostile shutdown) means loss of mint-backed ecash. Acceptable for working balances; not appropriate for treasury reserves. Mitigation: mint-diversification.
• Bears on: falsifier-relevant-to Stack-FA §8.1 CP2 / §8.2-S3 — specifically the falsifier condition “agents hold reserve-scale balances in ecash rather than operational-scale balances,” which would make mint trust systemic rather than scoped.

Record — Agent custody attack surface. (confirmed risk)

• Detail: software-managed keys controlled by autonomous agents introduce attack surfaces human-custodied keys do not: rogue agent behavior, key theft via prompt injection, treasury attacks, Sybil attacks on multi-agent settlements, social engineering against the humans operating agent infrastructure. The remote-signer architecture in lightning-agent-tools (signer holds keys, never connects to the public network) is the canonical mitigation; agent custody’s operational-security problem is not solved by substrate properties alone.
• Bears on: falsifier-relevant-to Stack-FA §8.1 CP4 / §8.2-S6 — deployment evidence of agents holding significant value under lighter primitives without loss-bounding, suffering systematic compromise, would trigger the S6 falsifier.

Stablecoin-substrate-side risk records.

Record — Issuer freeze surface. (measured risk)

• Detail: documented and exercised at scale (Circle / Tornado Cash, August 2022, ~$8.2M; Tether cumulative >$1B per attestations). A structural requirement of regulated-issuer licensing, not a bug to be patched.
• Bears on: supports Thesis-FA C2 and §8.1 CP1; supports Border-Zone-FA B4 (asset-side freeze fails Constraint 2). Quantitative record in §A.2.

Record — AgentCore stack custody layers. (confirmed risk)

• Detail: the deployed stack has multiple independent intermediary surfaces — Coinbase wallet/exchange custody discretion, Stripe payment-processor surface, Circle USDC freeze capability — each an independent intermediary-action surface under regulatory pressure.
• Bears on: supports Doctrine-FA §8.1 CP2 (the integration-scenario stack remains intermediated) and Border-Zone-FA B6; tests Doctrine-FA P6 (each intermediary surface is a freeze/KYC point that P6 predicts becomes a standing liability for agents under regulatory attention).

Record — Regulatory pressure trajectory. (confirmed risk; forward-looking)

• Detail: MiCA in the EU, ongoing US enforcement against stablecoin issuers, and sanctions-regime evolution are tightening — not loosening — the regulated-stablecoin operating environment across most jurisdictions in 2026. Integration-scenario use cases (content with issuer counterparty risk) are unaffected; parallel-economy use cases requiring censorship-resistance are increasingly disadvantaged on the competing substrate.
• Bears on: tests Doctrine-FA P6 (regulatory pressure routes autonomy-requiring commerce toward the censorship-resistant substrate — this record is the live evidence P6’s directional prediction will be evaluated against) and Doctrine-FA P3 (accommodation narrow, not broad); falsifier-relevant-to the P6 falsifier (tightening agent regulation instead consolidating agent commerce onto compliant intermediated rails without driving a meaningful share to censorship-resistant substrates).

Cross-substrate risk records (bridge-zone).

Record — Bridge counterparty risk at machine tempo. (confirmed risk)

• Detail: a bridge freeze during a high-frequency agent workflow has different consequences than during human-tempo transactions. Hot/cold treasury separation and multi-bridge redundancy are the architectural mitigations; deployed practice is still maturing.
• Bears on: supports Border-Zone-FA B3 (custodial-bridge identity/freeze surfaces at the boundary) and B5 (compliance at the gateway); supports Stack-FA S6 (hot/cold separation as the mitigation pattern).

Record — Conversion-mechanic attack surfaces. (confirmed risk)

• Detail: slippage attacks, MEV exposure during atomic swaps, oracle manipulation on bridge contracts — real for agents using cross-substrate conversion. Engineering attention from Boltz, Lightning Loop, and the broader DEX ecosystem; not yet a solved problem.
• Bears on: supports Border-Zone-FA conversion-mechanics treatment (submarine-swap and atomic-bridge templates); Stack-FA defers conversion mechanics to Border-Zone-FA by scope (S8).

Record — Jurisdictional shopping at scale. (confirmed risk)

• Detail: as agents and their custodians/bridges operate across jurisdictions, compliance-routing complexity grows. The compliance-at-the-gateway-boundary pattern handles this cleanly when implemented; sloppy architectures leak compliance into the protocol layer and compromise the parallel-system property.
• Bears on: falsifier-relevant-to Border-Zone-FA B5 (compliance-at-the-gateway preserves divergence only if the protocol layer stays unmodified) and Stack-FA S8 (scope boundary — compliance belongs at the bridge, not the substrate).

§B — Log (reverse chronological — newest first)

Dated event records on specific developments. Each: what happened / why it matters (structural significance, narrative stripped) / Bears on / epistemic tag / sources.

2026-05-07 — AWS Bedrock AgentCore Payments launches with Coinbase x402 + Stripe Privy (confirmed)

• Event. AWS announced Amazon Bedrock AgentCore Payments — infrastructure enabling autonomous agents to make real-time purchases using stablecoins. Build: Coinbase (x402 protocol on HTTP 402; Coinbase Agentic Wallets; compliance infrastructure) + Stripe (payment infrastructure and wallet integrations via Privy, acquired 2025). Settlement: USDC on Base, ~200ms confirmation, sub-cent per transaction. First version targets micropayments (APIs, data feeds, paywalled content). Enterprise customers at launch: Thomson Reuters, Warner Bros. Discovery, Cox Automotive, PGA TOUR.
• Structural significance. First Tier-1-enterprise production deployment of the integration scenario for agent payments. The customers are Fortune-500 enterprises operating in the regulated USD-denominated economy, not crypto-native early adopters. The stack serves the integration-scenario subset (USD-denominated, regulated-counterparty, issuer-counterparty-risk-acceptable) and the structural prediction is that it does so without absorbing the parallel-economy subset (agent activity requiring all four conjunctive constraints). The L402-vs-x402 naming convergence is the protocol-level expression of the substrate divergence: same status code, different settlement currencies, different trust models, two competing production stacks.
• Bears on: supports Border-Zone-FA B6 (use-case bifurcation) and Doctrine-FA §8.1 CP2 (integration scenario operationally deployed); tests Doctrine-FA P1 (which subset proves larger over 2–5 years) and Doctrine-FA P6 (the intermediated stack’s freeze/KYC/sanctions surfaces under agent regulation); supports Thesis-FA §8.1 CP1 (regulated stablecoins as substrate — operational confirmation).
• Sources. AWS: Agents that transact — Amazon Bedrock AgentCore Payments; The Block; CoinDesk; CryptoTimes. Structural treatment: Research/Border-Zone-Existing-Bridges.md §8; Research/Border-Zone-Competing-Substrate-Analysis.md CP1.

2026-05 — Routstr: Bitcoin-powered AI-inference marketplace (Cashu + Lightning + Nostr) (confirmed)

• Event. Routstr — open-source protocol + reference implementation (routstr-core, GPL-3.0; v0.4.3 May 2026) — runs a payment-gated reverse proxy in front of OpenAI-compatible LLM APIs plus a Nostr marketplace for provider discovery. Payment in Cashu ecash (token-as-API-key); provider earnings over Lightning; discovery/pricing as Nostr events. No accounts, KYC, or cards. HRF Top-15 Freedom Tech Project of 2025; supported under HRF “AI for Individual Rights.”
• Structural significance. The cleanest deployed instance of an agent buying a service on the Bitcoin payment stack (Cashu + Lightning) rather than the card/stablecoin stack — Cashu-token-as-API-key is a concrete bearer-credential answer to “how does an autonomous agent pay without a human-held account.” A Cashu-track instance: standardizes on Cashu (not Fedimint), bearer-token payment (not L402/NWC) — demonstrates one branch of the payment-tech stack, not all of it. Gap/collaboration opening: no llms.txt/agent-first surface.
• Bears on: supports Stack-FA S3 (Cashu bearer-ecash layer) + S4 (agent-integration: bearer credential) + S5 (deployed architecture); supports Doctrine-FA D3 (a deployed divergent instance, distinct from the incumbent stacks).
• Sources. Routstr; docs; GitHub: Routstr/routstr-core; HRF Top-15 Freedom Tech 2025.

2026-05 — Competing-substrate landscape broadens beyond AgentCore (AP2, Circle Nanopayments, Skyfire, x402 Foundation) (confirmed — landscape record)

• Event. (digests several 2025–26 developments.) Google AP2 (Agent Payments Protocol), launched September 2025: a 60+-organization consortium (Mastercard, American Express, PayPal, Coinbase, Adyen, Revolut, Worldpay, Salesforce, Intuit) with an A2A x402 extension built alongside Coinbase, the Ethereum Foundation, and MetaMask. x402 contributed to a dedicated x402 Foundation under the Linux Foundation (April 2026); 119M+ tx on Base. Circle Nanopayments (mainnet May 2026): gas-free USDC micropayments from $0.000001, x402-v2-compatible. Skyfire (“Agent Trust Stack,” backed by a16z CSX, Coinbase Ventures, Brevan Howard): Visa/Mastercard/Discover/USDC.
• Structural significance. The competing stack is plural and standardizing — at the governance layer (x402 Foundation; AP2 consortium), not just per-product. All standardize on stablecoins, card networks, and Ethereum/Solana, not Bitcoin (MetaMask, on the A2A x402 extension: “Ethereum will be the backbone”). This confirms Doctrine-FA D1’s mutual-exclusion mechanism: incumbents build a parallel stack preserving the issuer-controlled, freezable property bundle their licensing requires. (empirical concession — Constraint 3) Circle Nanopayments’ gas-free design narrows the sub-cent micropayment-economics gap on the stablecoin payments leg; it does not touch Constraints 1–2 (issuer freeze surface persists). (reality-check) CoinDesk (March 2026) noted x402 transaction demand remains thin relative to rail capacity — the substrate question is unsettled, not decided.
• Bears on: supports Doctrine-FA D1 + P1/P2/P5 (incumbents consolidate a competing-but-incumbent stack; substrate-selection precedes scale); supports Border-Zone-FA §8.1 CP1 roster update + B6 (use-case bifurcation); tests Thesis-FA §8.1 CP1 (regulated-stablecoin substrate — now plural deployment evidence).
• Sources. Google Cloud: Announcing AP2; GitHub: google-a2a/a2a-x402; x402.org; Coinbase x402 docs; Circle Nanopayments; Skyfire.

2026-04 — Lightspark Grid adds AI-agent bounded delegation (hybrid Lightning-rail stack) (confirmed)

• Event. Lightspark (led by ex-PayPal president David Marcus) added AI-agent bounded delegation to its Grid Global Accounts: agents get funded, scoped, auditable “pockets” with wallet-level spending limits, approved payees, per-transaction/daily/monthly caps, approval thresholds, and instant revocation. Grid settles over Lightning among multiple rails but is built on “Bitcoin and stablecoins” — branded USD/stablecoin accounts, Visa debit cards, instant Bitcoin conversion — and Lightspark has published an Agent Payments Protocol (AP2) vision aligning it with Google’s consortium.
• Structural significance. A Lightning-rail multi-rail product, not a Bitcoin-substrate one: the rail is Bitcoin’s; the asset (issuer-controlled USD/stablecoin) and the trust model (wallet provider + issuer + card network) are the incumbent’s. The instructive case in the competing roster because it is the closest to the substrate — a Lightning-native, Bitcoin-credentialed team still chose dollar/stablecoin denomination, card-network reach, and the AP2 stack. Confirms the divergence is about the asset and the trust model, not the rail. The bounded-delegation primitive (funded, scoped, revocable agent pockets) is a useful treasury-control pattern independent of the asset question.
• Bears on: supports Border-Zone-FA B4 (Lightning-rails-for-stablecoins ≠ Lightning-substrate) and B6 (use-case bifurcation; a hybrid actor serving the incumbent-economy subset over Lightning rails); supports Doctrine-FA D1 (the incumbent property bundle is preserved even by a Lightning-native builder); tests Doctrine-FA P6 (the intermediary layers — issuer, card network, wallet provider — are the freeze/KYC surfaces P6 predicts become a standing liability for agents under regulatory attention).
• Sources. Lightspark — Agent Payments Protocol (AP2); ITBrief: Lightspark adds AI agent controls to Grid; Bitcoin Magazine: Grid Global Accounts.

2026-03-21 — USDT live on Lightning via Taproot Assets (confirmed)

• Event. Tether CEO Paolo Ardoino confirmed USDT is live on Bitcoin’s Lightning Network via Lightning Labs’ Taproot Assets protocol, completing a 14-month integration begun at the Plan B Forum (El Salvador, January 30 2025). Bitfinex to issue USDT on Lightning per Tether. Follows the June 2025 Taproot Assets v0.6 release (“Bitcoin’s Decentralized FX Network”).
• Structural significance. A Lightning-rails bridge for the stablecoin, not a Lightning-substrate bridge. The issuer (Tether) retains freeze capability on its issuance regardless of which rail the asset moves over; Constraint 2 (censorship-resistance) still fails for the asset side even though rail-side properties (sub-cent, machine-tempo) are excellent. USDT-on-Lightning serves integration-scenario use cases and does not make stablecoins suitable as the parallel-economy substrate. The bridge changes the rail, not the asset.
• Bears on: supports Border-Zone-FA B4 (the exact rails-vs-substrate claim: rail-side passes Constraints 1/3/4, asset-side fails Constraint 2 by design); supports Thesis-FA §4 (substrate evaluation — regulated stablecoins row).
• Sources. Tether: USDt to Bitcoin’s Lightning Network; BTC.network fee-market analysis; Speed Wallet; Lightning Labs Taproot Assets v0.6. Operational treatment: Research/Border-Zone-Existing-Bridges.md §4.

2026-03 — Bitcoin Policy Institute publishes AI Models Overwhelmingly Prefer Bitcoin and Digital-Native Money Over Traditional Fiat (measured)

• Event. BPI published the study in March 2026. Method: 9,072 scenarios across 36 frontier language models, neutral scenario design (no leading prompts); each scenario asked the model to choose a preferred monetary instrument from a candidate set. Headline: Bitcoin the top overall monetary preference at 48.3% of responses, and dominant on the store-of-value dimension at 79.1%; over 90% favored digitally-native money over fiat (stablecoins led payment-preference scenarios at 53.2%). Per-provider the result was uneven — one provider’s models chose Bitcoin in 68% of responses, another’s in 26% — and the strongest single-model consensus anywhere in the study was 91.3%; wide spread, one-directional.
• Structural significance. Load-bearing empirical anchor for Thesis-FA C3 (substrate-preference signal) and Doctrine-FA P1 (substrate-selection-precedes-scale). Establishes that frontier models reasoning about substrate selection under neutral choice converge substantially toward Bitcoin without ideological prompting — consistent with the four-conjunctive-constraints argument. The study measures preference under inference, not deployed-flow dominance; convergent independent replication would strengthen the signal, contrary results would weaken it. No replication published as of May 2026.
• Bears on: supports / tests Thesis-FA C3 and §6; Doctrine-FA P1 and §7.2; falsifier-relevant-to Thesis-FA §8.2 (replication showing preference shift). KB origin: The AI-agent monetary substrate case § The empirical signal.
• Sources. Bitcoin Policy Institute — Study: AI Models Overwhelmingly Prefer Bitcoin and Digital-Native Money Over Traditional Fiat (March 3 2026). (BPI ai models prefer bitcoin research)

2026-02-11 — Lightning Labs releases lightning-agent-tools (confirmed)

• Event. Lightning Labs open-sourced lightning-agent-tools — a production AI-agent toolkit on the Bitcoin substrate. Seven composable skills: (1) running a Lightning node programmatically; (2) remote-signer key isolation; (3) baking scoped macaroons in five preset roles; (4) paying L402-gated APIs via lnget; (5) hosting paid endpoints via Aperture; (6) querying node state through MCP; (7) orchestrating end-to-end buyer/seller workflows.
• Structural significance. First Tier-1 production deployment of the Bitcoin-substrate agent-payment stack — the operational counterpart to the Thesis. Activates L402 (specified 2020) from “interesting protocol” to “production agent-commerce stack with deployed tooling.” Shipped February 2026; AWS AgentCore shipped May 2026 — the two competing-substrate production stacks emerged within 90 days of each other on directly comparable surfaces, making the Independence Doctrine’s prediction testable in real time.
• Bears on: supports Stack-FA S4 (integration primitives — L402, MCP, scoped credentials), S5 (deployed wallet architectures — the canonical reference implementation), S6 (security model — remote-signer isolation, scoped macaroons), and Thesis-FA C4 + §9; tests Doctrine-FA P1.
• Sources. Lightning Labs: The Agents Are Here and They Want to Transact (Feb 11 2026); Bitcoin Magazine, The Block, BitcoinEthereumNews coverage. Capability enumeration: Stack-FA §6; Thesis-FA §9.

Maintenance and refresh protocol

§A refresh cadence. At least quarterly, plus on any significant deployment shift: a new substrate stack going live; a major freeze incident; replication of the BPI study; substantial Lightning capacity or volume movement; an emerging protocol displacing a deployed pattern. Each refresh updates the section-A-last-refreshed frontmatter field and re-evaluates needs-refresh-tagged records (currently: Lightning capacity Q1–Q2 2026; Fedimint production federation counts).

§B append cadence. As developments warrant. Single dated records for specific events; multi-week composite records acceptable for slower-moving developments. Each record names the event, its structural significance, the canonical claim-IDs it bears on, and primary sources.

Defer-pattern (locked 2026-05-26). The canonical surfaces and their FA twins link out to this surface for ongoing empirical tracking rather than carrying it inline. The FA twins carry an inline empirical figure only where it sharpens a falsifier (currently: the Lightning-capacity figure in Stack-FA §8.1 CP1); all routine empirical updates defer here. When a canonical surface is tempted to cite a current number, the rule is: name it once, then defer to Field Notes.

References and provenance

Primary canonical source.

• Field Notes — the user-approved v0 canonical empirical-record surface; source for all records here. This document is the For-Agents twin.

Canonical claim-ID series this surface cross-references.

• Thesis-FA — substrate-selection argument; C1–C6 (esp. C2 censorship-resistance, C3 BPI anchor, C4 deployed system).
• Independence-Doctrine-FA — divergence doctrine; D1–D5 and predictions P1–P6 (esp. P1 substrate-selection-precedes-scale, P6 regulatory-pressure-feeds-not-threatens).
• Border-Zone-FA — bridge architecture; B1–B6 (esp. B3 protocol-vs-custodial bridges, B4 Taproot rails-vs-substrate, B5 compliance-at-the-gateway, B6 use-case bifurcation).
• Stack-FA — substrate architecture; S1–S8 and the §8 counter-positions/falsifiers (esp. S2/CP1 liquidity, S3/CP2 mint-trust, S6/CP4 agent custody).

Human-track canonical surfaces.

• Thesis, Independence-Doctrine, Border-Zone, Stack, The-Story.

KB origin.

• The case for investing in Bitcoin § AI-agent monetary substrate case; The AI-agent monetary substrate case (dedicated KB note for the four-constraints argument).

Date stamps. Document created 2026-05-31; last verified 2026-05-31. §A snapshot last refreshed 2026-05-26 (mirrors the canonical Field Notes; not re-measured this pass). Primary-source URLs in §B; metric sources named per §A record.