ARTICLES
Technology and business,
from hands-on practice.
Articles on AI agents, business systems, financial AI, and e-commerce.
We write from the perspective of designing, building, and operating technology in real business contexts.
CATEGORIES
ALL ARTICLES (PAGE 2 OF 7)
What Microsoft's Work IQ Expansion Means for Business Terms and Decision Criteria
Build on existing definitions, but check where they diverge from your own criteria. Work IQ's extension is a starting point for separating terms, procedures, and approvals.
AI Control Plane: Managing AI Agents Over Time
Deciding permissions is only the start. An agent register that tracks owners, models and configuration, actual settings, and retirement keeps AI agents manageable over time.
AWS and Salesforce's Zero Copy: What It Reduces and What You Still Need to Check
Zero copy reduces up-front replication, but some methods still store data temporarily. We set out five things to check before adoption, covering permissions, retained data, and how to stop access.
Exchange Inspector: Designing Verification for AI Components
A way to inspect community-built AI components before use has been announced. We look at connecting an external inspection result to your own usage conditions and ongoing review.
What Accenture's 1,000-Person FDE Plan Says About Running AI with a Small Team
A company that cannot staff a dedicated AI team still has to cover the same roles. Here is one way to divide them.
Agentforce's Seven Role-Specific Agents: Defining Permissions
Agents designed for specific business functions are a useful starting point for deciding what to delegate to AI — but their names and stated roles do not define the access permissions and approval requirements a company needs.
Meta Muse's Sentinel: Separating Execution from Authorization
Meta placed the authorization mechanism outside the executing agent. What that split, and the limits layered into its payment flow, suggest for enterprise AI agent design.
What Is GPT-6 Astra? Differences from GPT-5.6 Sol, Capabilities, Pricing, and Availability
What changed in GPT-6 Astra: a comparison with Sol, what it can do, which plans it is available on, what it costs, and what "Critical" actually means—as of September 9, 2026.
FINRA's Six AI Agent Risks and Supervisory Procedures
FINRA's six risks and its considerations, read apart from the obligations in Rule 3110—and carried through to what a written supervisory procedure actually has to say.