Prof Hernan Huwyler, MBA CPA

Learn how to build a production-ready policy engine for AI agents. This guide covers runtime governance, path-aware enforcement, progressive rollout strategies, and real code examples. Stop relying on prompts alone,enforce safety, compliance, and trust with architectural controls that actually work. https://hernanhuwyler.wordpress.com/2026/06/27/how-to-build-a-policy-engine-for-ai-agents-without-losing-control/?utm_source=mastodon&utm_medium=jetpack_social

Your Vendor’s “We Don’t Train On Your Data” Promise Is a Sentence, Not A Data Architecture Why the real exposure in generative, predictive, and agentic AI contracts lives in fine-tuning, logs, and retrieval, not in the one line everyone quotes back to legal Every procurement team has now heard the sentence. A vendor says it, a sales deck repeats it, and somebody on the buying side writes it into the approval memo as if it closes the risk. https://hernanhuwyler.wordpress.com/2026/07/21/your-vendors-we-dont-train-on-your-data-promise-is-a-sentence-not-a-data-architecture/?utm_source=mastodon&utm_medium=jetpack_social

Move past trusting model reasoning. The 2026 standard for agentic AI is consequence correctness. Stop hope-based automation and implement a runtime control chain that stops critical failures, like duplicate payments and prompt injection, before they hit your system of record. https://hernanhuwyler.wordpress.com/2026/07/28/the-seven-gates-every-ai-agent-must-clear-before-it-can-act-and-most-skip-at-least-three/?utm_source=mastodon&utm_medium=jetpack_social

Most AI security audits miss 70% of the attack surface. Training data, prompt pipelines, model weights, and agentic tool calls are #AI's real battleground. Free open-source #threatmodeling toolkit for engineers and architects: https://github.com/hwyler/ai-threat-modeling-toolkit https://hernanhuwyler.wordpress.com/2026/07/30/a-practical-guide-for-engineers-architects-and-governance-teams-who-need-to-get-it-right/?utm_source=mastodon&utm_medium=jetpack_social

Article 50 wasn't delayed. High-risk systems were. Your chatbot disclosure, content watermarking, and deepfake labels are due Sunday, August 2, 2026, with fines up to €15M or 3% of global revenue. #EUAIAct #AICompliance #AITransparency #Article50 #AIGovernance https://hernanhuwyler.wordpress.com/2026/08/03/the-eu-ai-acts-transparency-rules-just-went-live/?utm_source=mastodon&utm_medium=jetpack_social