B.E.R.R.: A Practical AI Framework

A Practical AI Framework For Learning Pros

Your manager drops into your office and says the organization is moving forward with AI on all enterprise platforms. You have dabbled with it on your own time, but when it comes to applying it to your actual work as a learning professional, you are not sure where to start. That is a reasonable place to be. Most L&D professionals are in the same position.

That is where B.E.R.R. comes in. I pronounce it Burrr, like a blast of cold air, because diving into AI without a road map can feel exactly that way. This framework is the coat.

Before we get into it, I want to reframe how you think about AI altogether. Stop thinking of it as a tool. Start thinking of it as a partner. I understand you are an expert in your field and can produce faster than you can type out a prompt. But the goal here is not to replace your expertise. It is to extend it. AI does not know your learners, your SMEs, or your organizational context the way you do. What it can do is help you move faster, go deeper, and pressure-test your thinking in ways that would otherwise require more time and more people.

According to a November 2025 study by Cornerstone OnDemand, 80% of employees now use AI at work. However, only 44% of workers have received any AI training at all, and just 16% received it consistently. As L&D professionals, that gap is both our challenge and our opportunity.

B.E.R.R. stands for Brainstorm, Enhance, Refine, and Review. These are four phases of content and work product development. The framework also gives you the language to explain your AI use to your team, your manager, or your executive stakeholders in terms that connect to efficiency and Return On Investment.

One important note: these phases do not have to be used in sequence and you do not have to use all of them every time. The framework is adaptable to your needs, your project, and your moment.

B: Brainstorm

Ideation is foundational to producing quality learning experiences. The best courses, curricula, and performance support tools usually start with a loose, generative conversation where good ideas and bad ideas get equal airtime. There was a time when L&D teams had the budget and bandwidth to spend days in a room with a whiteboard doing exactly that. Those days are largely gone. Budgets are tighter, teams are smaller, and timelines have compressed.

AI fills that gap. Think of it as a thought partner that never runs out of energy, never dominates the room, and never stops asking what else. Below are three prompts you can use progressively, moving from open exploration to a concrete direction. The scenario: you are designing a new onboarding curriculum for a hybrid workforce.

• Prompt 1 — Open Exploration
  I am designing an onboarding curriculum for a hybrid workforce at [type of organization]. I am in the early stages and want to explore a wide range of instructional approaches. Generate ten distinct learning design concepts that vary in format, modality, and learner engagement strategy. Include a one-sentence rationale for each.
• Prompt 2 — Narrow and Compare
  Based on the following three concepts I found most promising [paste concepts], compare them against these criteria: alignment with adult learning principles, feasibility for a hybrid delivery model, and scalability across departments. Identify the strongest option and explain why.
• Prompt 3 — Build It Out
  Take the winning concept and develop a high-level design brief that includes the learning objectives, suggested modalities, target audience considerations, and three potential module titles I can bring to my SMEs.

Tip

If brainstorming is a regular part of your workflow, build a reusable prompt template based on what worked. You can even ask AI to help you write it: "Help me write a reusable brainstorming prompt for [type of L&D project] that I can adapt for future use."

E: Enhance

You have your direction. Now the question is whether your content is performing at its full potential. Enhancement is about making your work sharper, more relevant, and more accessible to the learners who need it.

Start with personalization. What if you had a partner who could analyze learner survey results, LMS data, focus group transcripts, or performance metrics and then tell you exactly how your content is landing and where it is falling short? That is a real use case for AI. Feed it your data artifacts alongside your draft storyboard or script and ask it to flag gaps, suggest improvements, and identify language that is not connecting with your target audience. Feedback like content being too verbose or lacking real-world application examples becomes immediately actionable rather than sitting in a spreadsheet.

• Prompt — Learner Needs Analysis
  Here is a draft of [storyboard, script, or course outline] and here is [learner survey data, LMS performance data, or focus group notes]. Review both and identify: (1) places where the content does not reflect what learners have said they need, (2) gaps in coverage, and (3) specific language or example changes that would better connect with this audience.

For L&D professionals supporting an international workforce, AI can also help you make content translation-ready before it ever reaches a localization vendor. This is different from translation itself. It means identifying idiomatic expressions, culturally specific references, or complex sentence structures in the English version that will lose meaning or create friction when converted. Catching these early saves budget and revision cycles downstream.

• Prompt — Translation Readiness
  Review the following content and identify any phrases, idioms, or sentence structures that may not translate cleanly into [target language or region]. Suggest plain-language alternatives for each flagged item.

Even with thorough SME collaboration and research, content can have blind spots. AI can scan your storyboard or script for missing perspectives, underrepresented scenarios, or examples that would be stronger for a specific learner population. Use it to ask the questions your reviewers might not think to raise until the course is already in development.

R: Refine

One of AI's most underrated capabilities is bringing coherence to content built in pieces. When you are pulling from multiple SME sources, working across script drafts, or consolidating feedback from several stakeholders, things can start to feel disjointed. Refine is about making it whole.

The most practical application here is length and focus. If your script or storyboard has grown beyond your allotted seat time or word count, you do not have to make every cut yourself. Ask AI to help you decide what stays and what goes based on your stated learning objectives, then review its recommendations and make the final call.

• Prompt — Tighten To Seat Time
  I need to reduce this script from [current word count] words to [target word count] words without losing the core learning objectives. Identify the sections that add the least instructional value toward [stated objective] and suggest what can be cut or condensed.

Refine is also where tone and voice come together. If your content has been through multiple rounds of SME edits, AI can help you smooth out inconsistencies so the final product reads as one consistent voice. This is especially valuable for modular courses where multiple writers or reviewers have touched the content.

• Prompt — Voice Consistency
  The following content was developed with input from multiple SMEs and contributors. Review it for consistency in tone, voice, and sentence structure. Flag any sections that feel inconsistent with the rest and suggest revised versions that match the overall style.

You can also use this phase to check your content against a structural framework. If you want a module to follow a problem-solution narrative arc, a specific scenario-based structure, or a story framework, AI can tell you where the current draft deviates and what to do about it.

R: Review

Of all four phases, Review is the one most learning professionals reach for instinctively when they think about using AI. And yes, AI handles the mechanics of language well. Grammar, punctuation, sentence structure. Honestly, that is the easy part. The more valuable uses go further.

If your content reaches a broad or diverse learner population, ask AI to evaluate readability across two fronts: whether the content is accessible for learners whose first language is not English, and whether the overall reading grade level matches the audience you are actually designing for. Both checks matter and neither requires a separate accessibility review tool.

• Prompt — Readability And Accessibility
  Review the following content for readability across three dimensions: accessibility for learners whose first language is not English, ease of processing for learners who may engage with written language differently, and overall reading grade level. Identify the target grade level that fits [your audience] and flag anything that exceeds it. Suggest specific revisions throughout.

For training content specifically, one of the most powerful Review prompts is an objectives alignment check. After all the SME input, stakeholder feedback, and revision cycles, it is surprisingly easy for a course to drift from its original intent. AI can read your finished content alongside your original learning objectives and tell you where the alignment holds and where it has slipped.

• Prompt — Objectives Alignment
  Here are the learning objectives for this module: . Review the draft content and assess how well each section supports those objectives. Identify any sections that drift from the objectives or fail to deliver on them.

You can also use Review to scan for citation integrity, checking whether any passages closely mirror publicly available research or source material so you can properly attribute before publishing or deploying.

Conclusion

The idea behind B.E.R.R. is simple: learning professionals need a starting point. The conversation around AI in L&D often swings between excitement about the possibilities and anxiety about what it means for our roles. Neither helps someone who just needs to do their work better.

This framework uses concepts every L&D professional already understands. Brainstorming with stakeholders. Enhancing content for your audience. Tightening a script. Reviewing before it goes to development. None of that is new. What is new is the speed and depth at which a thoughtful AI partner can support each of those steps.

If you are working in an organization that is cautious about AI adoption, B.E.R.R. gives you a structured, defensible way to bring it in. It does not suggest the machine is doing your job. It makes clear that you are doing your job better, with better support. That has always been the job. B.E.R.R. just makes it easier to get there.