https://nicholasg3.github.io/ai-strategy/ AI Strategy is Nicholas Garcia's vertical on where AI should change coordination, judgment, learning speed, and commercial leverage — for founders, consultants, and C-suite operators. en-us https://nicholasg3.github.io/ai-strategy/posts/agents-need-memory-architecture.html strategy-publisher-issue-032-agents-need-memory-architecture Fri, 24 Jul 2026 14:00:00 GMT Research synthesis: production agents store experience in prompt rules and model weights that drift apart. Field analysis of 1,990 public discussions yields a three-episode memory architecture process. https://nicholasg3.github.io/ai-strategy/posts/ai-disclosure-clock.html strategy-publisher-issue-031-ai-disclosure-clock Thu, 24 Jul 2026 13:40:00 GMT News digest: EU AI Act Article 50 transparency duties apply 2 August 2026; Commission guidelines of 20 July make disclosure a two-week operator deadline. https://nicholasg3.github.io/ai-strategy/posts/ai-coverage-gap.html strategy-publisher-issue-030-ai-coverage-gap Thu, 23 Jul 2026 11:20:00 GMT Generative AI insurance exclusions turn residual risk into an operator ledger: exposure class, policy line, exclusion surface, residual owner, control evidence, and stop rule. https://nicholasg3.github.io/ai-strategy/posts/ai-browser-boundary.html strategy-publisher-issue-029-ai-browser-boundary Thu, 23 Jul 2026 10:50:00 GMT Agentic browsers turn a reading surface into an execution surface: session privilege, instruction channels, tool reach, data boundaries, work classes, and incident paths. https://nicholasg3.github.io/ai-strategy/posts/ai-source-rights-ledger.html strategy-publisher-issue-028-ai-source-rights-ledger Wed, 22 Jul 2026 12:45:00 GMT AI search is unbundling the old crawler bargain into access, citation, answer, training, payment, and relationship rights. https://nicholasg3.github.io/ai-strategy/posts/ai-control-gap-ledger.html strategy-publisher-issue-027-ai-control-gap-ledger Wed, 22 Jul 2026 12:30:00 GMT Enterprise AI accountability is outrunning visibility, action authority, financial exposure, incident response, and business ownership. https://nicholasg3.github.io/ai-strategy/posts/ai-audience-ledger.html strategy-publisher-issue-026-ai-audience-ledger Wed, 22 Jul 2026 12:10:00 GMT AI search turns founder publishing into an audience ledger: crawl access, answer presence, referral quality, owned relationships, conversion rights, and proof loops. https://nicholasg3.github.io/ai-strategy/posts/ai-slop-audit.html strategy-publisher-issue-025-ai-slop-audit Wed, 22 Jul 2026 11:50:00 GMT AI-assisted work should be audited for accuracy, specificity, judgment, review burden, and learning before generated volume is mistaken for operating leverage. https://nicholasg3.github.io/ai-strategy/posts/ai-memory-ledger.html strategy-publisher-issue-024-ai-memory-ledger Wed, 22 Jul 2026 11:35:00 GMT AI memory should be governed through a ledger of preference, role, project, sensitive, and institutional context before persistent personalization becomes hidden policy. https://nicholasg3.github.io/ai-strategy/posts/ai-model-mix-ledger.html strategy-publisher-issue-023-ai-model-mix-ledger Wed, 22 Jul 2026 11:25:00 GMT AI buying should be managed through a ledger of frontier APIs, open-weight models, hosted models, vendor bundles, specialist models, and no-model boundaries. https://nicholasg3.github.io/ai-strategy/posts/ai-productivity-gap.html strategy-publisher-issue-022-ai-productivity-gap Wed, 22 Jul 2026 11:05:00 GMT AI productivity should be judged through a ledger of adoption, daily use, workflow redesign, measurement, learning, and capital discipline. https://nicholasg3.github.io/ai-strategy/posts/ai-distribution-boundary.html strategy-publisher-issue-021-ai-distribution-boundary Wed, 22 Jul 2026 10:20:00 GMT AI search, workplace agents, and browser agents should be managed as a distribution boundary: discovery, evidence, action, context, relationship, conversion, and feedback. https://nicholasg3.github.io/ai-strategy/posts/agent-identity-boundary.html strategy-publisher-issue-020-agent-identity-boundary Wed, 22 Jul 2026 10:10:00 GMT AI agents should be governed through an identity boundary: actor, authority source, session, tool scope, human checkpoint, record, and revocation. https://nicholasg3.github.io/ai-strategy/posts/ai-disclosure-ledger.html strategy-publisher-issue-019-ai-disclosure-ledger Wed, 22 Jul 2026 09:50:00 GMT AI transparency should be managed as a disclosure ledger: surface, audience, origin, control, and evidence. https://nicholasg3.github.io/ai-strategy/posts/ai-evaluation-gap.html strategy-publisher-issue-018-ai-evaluation-gap Wed, 22 Jul 2026 09:30:00 GMT AI evaluation should become a live evidence loop: task fit, workflow fit, failure map, cost burden, and learning cadence. https://nicholasg3.github.io/ai-strategy/posts/ai-agent-permission-ledger.html strategy-publisher-issue-017-ai-agent-permission-ledger Wed, 22 Jul 2026 09:10:00 GMT Agentic AI should be governed through a permission ledger: read, reason, write, spend, message, change, owner, audit trail, and stop rule. https://nicholasg3.github.io/ai-strategy/posts/ai-work-cadence.html strategy-publisher-issue-016-ai-work-cadence Wed, 22 Jul 2026 08:52:00 GMT AI adoption should be governed as a work cadence: artifact, rhythm, handoff, review burden, and job redesign. https://nicholasg3.github.io/ai-strategy/posts/ai-token-budget.html strategy-publisher-issue-015-ai-token-budget Wed, 22 Jul 2026 08:36:00 GMT Inference should be governed as a token budget: model tier, context budget, agent loop, value class, and owner. https://nicholasg3.github.io/ai-strategy/posts/ai-productivity-receipt.html strategy-publisher-issue-014-ai-productivity-receipt Wed, 22 Jul 2026 08:18:00 GMT AI benchmark gains should become business commitments only after the company can show task, workflow, economic, and trust receipts. https://nicholasg3.github.io/ai-strategy/posts/founder-distribution-ledger.html strategy-publisher-issue-013-founder-distribution-ledger Wed, 22 Jul 2026 08:02:00 GMT Treat founder publishing as a four-column ledger — shelf, relationship, conversion, proof — so channel fashion does not masquerade as strategy. https://nicholasg3.github.io/ai-strategy/posts/asean-market-entry-test.html strategy-publisher-issue-012-asean-market-entry-test Wed, 22 Jul 2026 07:35:00 GMT ASEAN digital integration should be treated as a market-entry test: first buyer, operating reuse, trust bridge, regulatory drag, and regional option value. https://nicholasg3.github.io/ai-strategy/posts/stablecoin-distribution-test.html strategy-publisher-issue-011-stablecoin-distribution-test Wed, 22 Jul 2026 07:20:00 GMT Stablecoin payments should be judged as a distribution test: buyer access, settlement value, product fit, and operating control. https://nicholasg3.github.io/ai-strategy/posts/ai-evidence-packet.html strategy-publisher-issue-010-ai-evidence-packet Wed, 22 Jul 2026 07:00:00 GMT AI procurement should move from demo judgment to evidence judgment before agents, copilots, and automation layers enter production workflows. https://nicholasg3.github.io/ai-strategy/posts/ai-payback-ledger.html strategy-publisher-issue-009-ai-payback-ledger Wed, 22 Jul 2026 06:41:00 GMT AI capex should be judged by asset, obligation, payback owner, proof milestone, and stop rule before spending scales. https://nicholasg3.github.io/ai-strategy/posts/ai-herding-problem.html strategy-publisher-issue-008-ai-herding-problem Wed, 22 Jul 2026 06:12:00 GMT When AI makes analysis cheap and crowded, executives need a clear doctrine for which strategic decisions remain human-owned. https://nicholasg3.github.io/ai-strategy/posts/ai-quality-test.html strategy-publisher-issue-007-ai-quality-test Wed, 22 Jul 2026 05:50:00 GMT AI spending should be judged as quality capital, a learning option, or theme exposure before funding expands. https://nicholasg3.github.io/ai-strategy/posts/ai-data-center-veto.html strategy-publisher-issue-006-ai-data-center-veto Wed, 22 Jul 2026 05:34:00 GMT Power, permitting, water, ratepayer politics, and capital discipline are becoming board-level constraints on AI strategy. https://nicholasg3.github.io/ai-strategy/posts/ai-strategy-theatre-test.html strategy-publisher-issue-005-ai-strategy-theatre-test Wed, 22 Jul 2026 05:02:00 GMT A five-part executive screen for separating strategic AI workflow change from demos, activity metrics, and unmanaged theatre. https://nicholasg3.github.io/ai-strategy/posts/ai-management-layer.html strategy-publisher-issue-004-ai-management-layer Wed, 22 Jul 2026 04:50:00 GMT Enterprise AI is being folded into managed work, agent registries, observability logs, provenance rules, licensing, and transparency obligations. https://nicholasg3.github.io/ai-strategy/posts/ai-vendor-warning-label.html strategy-publisher-issue-003-ai-vendor-warning-label Wed, 22 Jul 2026 00:20:00 GMT Agent-security guidance is becoming the AI industry's shared-responsibility model: a practical safety manual today and a future control record after the first serious incident. https://nicholasg3.github.io/ai-strategy/posts/ai-workflow-investment-screen.html strategy-publisher-issue-002-ai-workflow-investment-screen Tue, 21 Jul 2026 09:45:00 GMT Companies are not short of AI use cases. They are short of managerial judgment about where autonomy should enter the business and where it should stop. https://nicholasg3.github.io/ai-strategy/posts/ai-strategy-doctrine-before-tools.html strategy-publisher-issue-001-ai-strategy-doctrine-before-tools Tue, 21 Jul 2026 09:22:00 GMT Founders do not have an AI tooling problem first. They have a doctrine problem.