Writing · Nº 041065 essays
The writing
Ideas from inside the work. Explore technology, entrepreneurship, and the practice of building things that last.
93 essays found · Page 2 of 8
Clear filtersStrategyExplainability vs Accuracy: What AI Buyers WantExplainability vs accuracy is a real tradeoff, and serious AI buyers often pick explainability. Here is why a decision you can prove beats one you cannot.StrategyGovernance Is the Moat Nobody Wants to BuildClaims discipline, consent records, audit trails, and human checkpoints are unglamorous. That is exactly why they make a durable competitive moat.OperatingHow to Deploy an AI Agent in a Regulated WorkflowDeploying an AI agent in a regulated workflow is about evidence, not capability. Here is how to make an agent's actions defensible to an auditor or a regulator.StrategyBuild a Governance Evidence Pack for ProcurementEnterprise procurement asks the same governance questions every time. Build an evidence pack once and answer them in a day instead of a stalled quarter.StrategyModel Risk Management: Selling AI to Banks Under SR 11-7Banks govern AI as model risk under SR 11-7. If you sell to them, your governance must survive validation. Here is what model risk management demands of vendors.StrategyWho Is Liable When Your AI Agent Gets It WrongWho is liable when an AI agent makes a mistake? The answer depends on whether you can prove what it did. Governance decides where the liability lands.ThesisOne Governed Foundation Under Every CompanyNineteen ventures share one governed substrate. Identity, audit trails, and guardrails are built once and inherited everywhere, not rebuilt per app.ThesisHow to Rebuild Trust After Your AI Fails a CustomerYour AI will fail a customer eventually. How to rebuild trust after an AI failure with honest disclosure, a real fix, and evidence it will not repeat.StrategyWhat HIPAA Buyers Need From an AI Vendor's GovernanceSelling AI into healthcare means clearing a governance bar most vendors miss. Here is what HIPAA-covered buyers actually require before they let your model near PHI.StrategyEnterprise Assurance for Selling AI to BanksSelling AI to a bank means clearing model risk management and third-party risk review. Here is what enterprise assurance looks like in financial services.OperatingVerifiable Voting for HOAs, Co-ops, and ClubsSmall communities run high-stakes votes on paper and email and cannot prove the result. Here is how verifiable voting fixes elections for HOAs, co-ops, and clubs.EngineeringLeast Privilege for AI Automation IntegrationsAn AI workflow with full access to every system is a breach waiting to happen. Here is how to apply least privilege so automation can only touch what it needs.