Praxis Creative Studio

We are currently witnessing a massive friction point in the L&D industry.

On one side, you have the promise of AI-generated content—synthetic avatars, automated narration, and rapid-fire video production. The efficiency gains are undeniable.

On the other side, you have the corporate realities of compliance, brand stewardship, and accessibility standards (Section 508, anyone?).

For many L&D teams, this friction has led to “AI paralysis.” They see the potential, but they fear that letting go of the manual creation process means letting go of their quality assurance.

They aren’t wrong—if you treat AI as an autonomous content generator, you will lose compliance.

But if you treat AI as a junior team member that requires an “Architect” to supervise, you unlock massive scalability.

Enter the “Human-in-the-Loop” (HITL) Model.

Scaling AI content in a corporate environment isn’t about generating more content; it’s about establishing the right governance. To do this effectively, I’ve been shifting my clients from “Content Creators” to “Content Architects.”

Here is how to maintain compliance while scaling:

1. Standardize the Inputs, Not Just the Outputs

AI hallucinations often stem from bad prompting. If you want compliant output, you need rigid input structures. Create a library of “Compliance-Approved Prompts” for your SMEs. Instead of giving them a blank text box, provide a structured template that forces the inclusion of brand-aligned messaging and accessibility considerations from the start.

2. The “Human Gatekeeper” Workflow

AI should never be the final step in your pipeline. Your workflow should look like this:

• Strategy: The human (you) sets the learning objective.
• Generation: The AI (e.g., Articulate, HeyGen, Synthesia) handles the visual assembly.
• Validation: The SME (Human-in-the-loop) verifies the technical accuracy and compliance.

This human verification step is not a bottleneck; it is your quality control. By placing a Subject Matter Expert in the review chair, you offload the historically heavy lifting of “production” to the AI while keeping the “authority” with your team.

3. Compliance by Design

Don’t retroactively check for accessibility. Embed it into your AI workflows. Use AI tools to generate transcripts or descriptive text that are WCAG-compliant before they are integrated into your Rise or Storyline modules. By treating compliance as a default setting rather than an afterthought, you eliminate the risk of late-stage rework.

The future of L&D isn’t about choosing between speed and standards. It’s about building an architecture that allows for both.

AI doesn’t replace the instructional designer. It frees us to spend less time on pixel-pushing and more time on high-level strategy—which is exactly where our value lies.

Are you building guardrails into your AI workflows, or are you still relying on manual review at the very end of your project lifecycle?

I’d love to hear how your teams are navigating this shift.