Creating an AI-automated workflow using content design principles

Content designers are increasingly asked to “use AI,” but far fewer are invited into the design of AI tools themselves. This talk walks through the real, messy, behind-the-scenes work of designing an industry-first AI-automated content generation workflow from Jira to production that’s built to support content designers without replacing their judgment, voice, or craft.

You’ll see how our team translated content design principles into an AI MVP, including how we defined agent roles, set quality standards, handled evaluation, and navigated trust, accuracy, and workflow fit. This isn’t a product demo or a hype tour: it’s a candid case study of what worked, what didn’t, and what we learned about designing AI systems that respect content designers as experts.

In this session, you’ll learn how to:

• Explain how content design principles can be applied to AI system design, not just outputs.
• Identify practical patterns for defining AI “roles” and boundaries in a writing workflow.
• Evaluate AI-generated content using criteria like accuracy, clarity, and relevance.
• Recognize common failure modes when AI tools ignore real content workflows.
• Apply lessons from our MVP to your own experiments with AI at work.