--- title: 14 Human Oversight Patterns for AI Systems | daily.dev description: Studying AI regulations, customer systems and its own services, Akka identified 14 recurring human oversight patterns for AI systems. Researchers found 206... canonical: https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: 14 Human Oversight Patterns for AI Systems | daily.dev og:description: Studying AI regulations, customer systems and its own services, Akka identified 14 recurring human oversight patterns for AI systems. Researchers found 206... og:url: https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy og:image: https://api.daily.dev/og/posts/Iuce5Gnzy.png og:image:alt: 14 Human Oversight Patterns for AI Systems og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # 14 Human Oversight Patterns for AI Systems **[Lightbend](https://daily.dev/sources/lightbend)** · 40 min read · 2 upvotes · 0 comments ## Summary Studying AI regulations, customer systems and its own services, Akka identified 14 recurring human oversight patterns for AI systems. Researchers found 206 human controls across a regulatory corpus of 1,751 controls from 241 regulations, and nearly all fit one of the 14 patterns. Each pattern defines who decides, the deadline, what happens if nobody decides, the allowed decisions and the record kept. A review of 34 AI governance and agent products found only three that turn a monitoring signal into an owned, recorded work item. The patterns will soon be available as templates in the Akka Agentic AI Platform. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What are the main categories of human-in-the-loop patterns required by AI regulations? Fourteen recurring patterns cover human oversight obligations found across AI regulations, including action approval, decision review, change authorization, exception, match and adjudication, outcome reconciliation, contest and appeal, override, flag and feedback, triage, hand-off, halt and restart, attestation, and oversight quality. Analysis of a corpus of 1,751 regulatory controls found 206 requiring human action, with 183 fitting cleanly into these 14 patterns. _Developers designing AI compliance workflows can track pattern catalogs like this on daily.dev._ ### What does Article 14(4)(d) of the EU AI Act require regarding human oversight of AI systems? It requires that the person overseeing a high-risk AI system can 'disregard, override or reverse the output of the high-risk AI system.' This provision underlies the 'override' oversight pattern, where a person sets aside or replaces a system output and records a reason, with the original output kept alongside the replacement for audit purposes. _Teams building EU AI Act compliance features can follow implementation details like this on daily.dev._ ### How long should a system wait before escalating an unapproved AI action to a more senior reviewer? In one implemented example, an unanswered payment approval escalates to the head of operations after 30 minutes, and the payment is canceled if nobody approves it within two hours. The default claim length for a reserved review event is 30 minutes, and events are marked 'at risk' once 75% of the time to deadline has passed. _Developers tuning escalation timeouts for approval workflows can compare patterns like this on daily.dev._ ## Similar posts on daily.dev - [The Inspector's Instinct](https://daily.dev/posts/the-inspector-s-instinct-v6oqv5ybh) · Webframp · 0 upvotes · 0 comments - [Stop Reviewing Your AI Agents. Start Governing Them.](https://daily.dev/posts/stop-reviewing-your-ai-agents-start-governing-them--vyihoxgkb) · Security Boulevard · 2 upvotes · 0 comments - [The Agentic AI Handbook: Production-Ready Patterns](https://daily.dev/posts/the-agentic-ai-handbook-production-ready-patterns-zlhrkqtw7) · Hacker News · 1 upvotes · 0 comments - [AI governance is now an engineering problem](https://daily.dev/posts/ai-governance-is-now-an-engineering-problem-p4jterxae) · LeadDev · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#ai-governance](https://daily.dev/tags/ai-governance), [#akka](https://daily.dev/tags/akka) [View this post on daily.dev](https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"14 Human Oversight Patterns for AI Systems","url":"https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy"},"datePublished":"2026-10-31T17:40:07.137Z","dateModified":"2026-10-31T17:40:36.861Z","description":"Studying AI regulations, customer systems and its own services, Akka identified 14 recurring human oversight patterns for AI systems. Researchers found 206...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/be67b33284bfa7312900f65107b7e518?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/be67b33284bfa7312900f65107b7e518?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Lightbend","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Lightbend","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/lightbend","url":"https://daily.dev/sources/lightbend"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,ai-governance,akka","timeRequired":"PT40M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Lightbend","item":"https://daily.dev/sources/lightbend"},{"@type":"ListItem","position":3,"name":"14 Human Oversight Patterns for AI Systems"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/14-human-oversight-patterns-for-ai-systems-iuce5gnzy#faq","mainEntity":[{"@type":"Question","name":"What are the main categories of human-in-the-loop patterns required by AI regulations?","acceptedAnswer":{"@type":"Answer","text":"Fourteen recurring patterns cover human oversight obligations found across AI regulations, including action approval, decision review, change authorization, exception, match and adjudication, outcome reconciliation, contest and appeal, override, flag and feedback, triage, hand-off, halt and restart, attestation, and oversight quality. Analysis of a corpus of 1,751 regulatory controls found 206 requiring human action, with 183 fitting cleanly into these 14 patterns. Developers designing AI compliance workflows can track pattern catalogs like this on daily.dev."}},{"@type":"Question","name":"What does Article 14(4)(d) of the EU AI Act require regarding human oversight of AI systems?","acceptedAnswer":{"@type":"Answer","text":"It requires that the person overseeing a high-risk AI system can 'disregard, override or reverse the output of the high-risk AI system.' This provision underlies the 'override' oversight pattern, where a person sets aside or replaces a system output and records a reason, with the original output kept alongside the replacement for audit purposes. Teams building EU AI Act compliance features can follow implementation details like this on daily.dev."}},{"@type":"Question","name":"How long should a system wait before escalating an unapproved AI action to a more senior reviewer?","acceptedAnswer":{"@type":"Answer","text":"In one implemented example, an unanswered payment approval escalates to the head of operations after 30 minutes, and the payment is canceled if nobody approves it within two hours. The default claim length for a reserved review event is 30 minutes, and events are marked 'at risk' once 75% of the time to deadline has passed. Developers tuning escalation timeouts for approval workflows can compare patterns like this on daily.dev."}}]} ```