Human-Centered Architecture01
3 Things AI: The “Start With the Moment” Edition
Three fresh signals show when to stop an expensive test, why customer corrections belong in evaluation, and where urgent AI work should run.
Back to journal
Twice-weekly quick read
Tuesday and Thursday
4–6 minute read
3 Things AI
Three signals. Three decisions.
A concise read connecting three AI shifts to business consequences and one practical operating move.
Series archive
Published editions
Human-Centered Architecture02
3 Things AI: The “Help Without Taking Over” Edition
Three fresh signals show how to reuse expert judgment, support a meeting without interrupting it, and measure what an AI agent actually remembers.
Decision Architecture03
3 Things AI: The “Test the Fit, Then Trust the Result” Edition
Three fresh studies suggest better tests for choosing workplace tasks, reviewing AI scores, and checking facts about less familiar companies.
AI Operating Models04
3 Things AI: The “Don’t Let Convenience Choose” Edition
Three fresh studies suggest practical tests for preserving staff learning, presenting choices to shopping agents, and keeping critical rules through memory compaction.
AI Operating Models05
3 Things AI: The “Is It Actually Finished?” Edition
Three fresh studies suggest practical checks for bilingual review, self-review costs, and using AI PCs as a small inference fleet.
AI Operating Models06
3 Things AI: The “Look Past the Best Answer” Edition
Three fresh studies offer practical tests for multi-step handoffs, generated possibilities, and faster business search without hidden losses.
Decision Architecture07
3 Things AI: The “Give the Task What It Needs” Edition
Three fresh studies suggest better tests for long context, visual business ideas, and AI systems that know when to defer.
Decision Architecture08
3 Things AI: The “Make the Work Checkable” Edition
Three new studies suggest practical tests for analytics answers, supply-chain risk signals, and the judgment people retain after AI assistance.
Decision Architecture09
3 Things AI: The “Keep the Decision Visible” Edition
Three new studies offer practical lessons on multilingual model choice, uncertainty in forecasts, and checking whether an image is synthetic.
Agent Systems010
3 Things AI: The “Real Work Has Consequences” Edition
Three new studies show why AI must be tested on completed accounting work, team dynamics, and the full security lifecycle of persistent memory.
Decision Architecture011
3 Things AI: The “More Is Not a Control” Edition
Three current signals show why more compute, more AI reviewers, and more confident prompts still require measured operating controls.
Decision Architecture012
3 Things AI: The “Boundaries Need Receipts” Edition
Three current signals show why agent containment, vendor portability, and sovereign AI claims now need evidence that survives contact with operations.
Canadian AI Governance013
3 Things AI: The “Controls Have to Reach Production” Edition
Three current signals show why AI transparency, patch ownership, and grounded simulation now belong in the operating system—not the policy binder.
AI Operating Models014
3 Things AI: Visibility, Workforce Repricing, Operational Integration — Decisions for Canadian SME Leaders
Three signals July 12–14, 2026 that change what Canadian SMEs must measure, how they price people, and how they embed AI into operations — with concrete owner-level moves.
Agent Systems015
Three signals that separate governed systems from prompt habits
Three near-term signals — context ownership, evaluation evidence, and outcome measurement — and what Canadian SMEs must change to move from prompt-first habits to governed AI systems.
AI Operating Models016
Three things AI: portability, behaviour disclosure, and cyber leadership — operational resilience for Canadian SMEs
Three current signals (June 30–July 2, 2026) that change how SMEs must design AI operations for resilience: model portability, mandatory behaviour disclosure mechanics, and shifting cyber leadership in standards bodies.
Agent Systems017
Model routing, agent governance, and Canada funding: 3 operating moves for SME operators
Three linked signals—cost-aware model routing, built-in agent control planes, and accelerating Canadian funding—translate into concrete SME operating decisions and owners.
Agent Systems018
Three operational signals: AI cost controls, agent permission boundaries, and machine-readable business trust
Three short, operational signals Canadian SMEs must act on now: new cost-controls from cloud providers, runtime enforcement for agent permissions, and emerging machine-readable credentials for vendor trust.
AI Operating Models019
3 Things AI: The Control Layer Wins
Three current AI signals translated into practical consequences for cost, permissions, and machine-readable trust.