Next Level Strategies For Implementing an HR/L&D AI Solution

In the Blanchard Innovation Lab, which I’m a part of as Director of Digital, we explore all sorts of things related to learning design and development—including generative and conversational artificial intelligence (AI). And just like you, we’ve seen a wide variety of new course and content creation generators pop up since it took off. Every day, new tools flood the market to help people create learning content, most of which is instructionally sound, but sometimes not as much.

At Blanchard, as we continue to explore new AI possibilities in the Innovation Lab, we are also developing and clarifying best practices to effectively work with this new technology. Here are four strategies that guide our planning. See how these principles could be put to work in your exploration of AI possibilities.

Discovery: Measure Twice, Cut Once

Discovery is the first critical step along a solution path. This stage is all about digging deep to uncover the pain points and challenges being faced. Think of it as laying the groundwork for a house: without a solid foundation, anything you build on top will be less stable. By thoroughly understanding the specific needs of an end user, we can tailor our AI solutions to address these issues effectively. This phase emphasizes the age-old wisdom of "measure twice, cut once," underscoring the importance of meticulous planning and preparation. It's about asking the right questions, listening carefully to the answers, and ensuring that we have a comprehensive understanding of the problems we're trying to solve.

There are three parts to successful discovery.

  1. Schema creation. In this phase you’re understanding the business issue, challenge, or need that you’re solving for in ways that will improve efficiency, quality, or ROI.
  2. Complex requirement gathering. To properly build out that schema or framework, you have to know what the people using the Generative Pre-trained Transformer (GPT) need and want vs. what is possible. Part of this requirement-gathering stage is about alignment. Not only are you refining your schema, you’re also aligning stakeholders’ expectations about what the end product or tool will be able to accomplish.
  3. Detailed information gathering. To ensure that your GPT results are accurate and comprehensive, you must access all available sources and data. The information it might need to access could range from something simple, like articles or small content pieces, to something extremely complicated, like connecting to a financial or customer database and getting real-time financial data. Be comprehensive but also discriminating in this phase. Some of the information you’ll receive will be useful and other parts will be irrelevant. A clear schema will make it easier to sift through the need to have, nice to have, and don’t need buckets.

Solution Design: Crafting Tailored AI Solutions

Once you’ve identified the challenges and pain points, it’s time to move on to the solution design phase. This is where creativity meets technology. At Blanchard, our team of experts designs AI solutions that are not only innovative but also perfectly tailored to meet the needs identified during the discovery phase. This bespoke approach ensures each solution is as effective as possible, addressing the specific pain points of each department with precision.

For example, in a corporate setting, imagine an HR department is grappling with an inefficient recruitment process. The team gets started on the solution design phase using insights from a successful discovery phase. They develop a customized AI recruiting tool to streamline talent acquisition. This AI tool proves effective at parsing through applications and identifying ideal candidates by learning from historical hiring data—and greatly improves the company’s recruitment strategy, making it vastly more efficient. This is the essence of generative AI design: merging creativity with technology to craft solutions that precisely address department-specific challenges.

Prompt Engineering: The Heart of Custom AI

You hear a lot about prompt engineering—and for a good reason. Prompt engineering is at the core of AI solutions. This phase involves setting up specific AI prompts that will guide the artificial intelligence in performing its tasks efficiently and effectively. It's a nuanced process that requires a deep understanding of both the technology and the unique challenges it aims to solve. The right prompts can mean the difference between a solution that's merely functional and one that truly transforms the way a department operates.

I like to use the metaphor of a genie in a bottle. You rub the bottle and a genie appears. Now you need to make an extremely specific and wise request. If you don't, you might not get what you want—or something unintended. It's the same with AI. Ask better questions, get better results.

Testing and User Feedback: Refining the Experience

The final stage of a good implementation plan involves rigorous testing and gathering user feedback. This is where we put our AI solutions to the test, in real-world scenarios, to ensure they perform as intended. User feedback is invaluable at this stage, providing insights that help us refine and improve our solutions. It's a process of continuous improvement, where feedback loops play a crucial role in ensuring our AI solutions not only meet but exceed the expectations of our users.

Key stakeholders who will use the GPT every day should test and review the output to ensure it's generating exactly what they want. If everything is successful, the next step is to make sure all users are upskilled, follow the prompt engineering, and understand how to get the magic they need out of the tool that’s been created.

It's helpful to remember every piece of software comes with bugs. If there weren't bugs, there would be no need for constant maintenance and code improvement. Humans are imperfect, and so is the code we create.

Making AI Work for Our Businesses

Now that we’ve covered some of our overall best practices, let’s take a look at how we make AI work specifically in an L&D setting.

  1. Clarify objectives. Understand what you want to achieve with AI in L&D. Are you looking to streamline design work, support video script writing, role play feedback, translate content into other languages, or something else? Understand the purpose of AI integration and who will be using it, such as instructional designers, learners, or writers.
  2. Integrate with human efforts. Determine how AI can complement human tasks rather than replace them, ensuring a partnership that leverages the strengths of both. AI should not replace the human element that is central to leadership training; it should complement it. If you use a tool to create a video script for a course, ensure a human is there to review and verify relevancy, accuracy, and cohesion.
  3. Train your team. Ensure your L&D team understands how to work with AI, interpret its outputs, and apply insights effectively.
  4. Iterate and improve. Use AI's feedback loops to continually refine your L&D initiatives. Remember, AI thrives on data, so the more you use it, the better it gets.
  5. Focus on the human experience. Keep the learner's journey at the heart of your AI strategy. In all AI endeavors, the learner's experience should remain paramount. AI in L&D isn’t just about smarter learning; it's about deeper, more meaningful development experiences.

Blanchard Is Looking Ahead and Beyond

At Blanchard we recognize the fusion of AI into our product development work is not just an upgrade—it's a revolution.

In adopting a new technology like generative AI, L&D professionals have new ways to elevate their skillset and focus on the needed consultative skills to solve complex problems that require training. AI will also serve as our ally in shaping a world where leadership is for the many, not the few, and where the potential of every individual is realized. AI increases accessibility, where amazing content can be tapped into by anyone. That’s part of our mission at the Blanchard Innovation Lab.

To learn more about our research into AI and other areas we are exploring at the forefront of human-centric technology forward designs, contact us. We’d love to discuss our latest projects—and learn about yours!

Editor's Note: 

Would you like to learn more about AI applications for learning and development? Download our free eBook, An HR / L&D Primer on Generative AI. You’ll find this post, together with five other perspectives from Blanchard thought leaders exploring the challenges leadership, learning, and talent development professionals are facing in today’s changing environment. Use this link to download your complimentary copy now.

About the Author

Romero Habib

Romero Habib is the Director of Digital for the Blanchard Innovation Lab, spearheading AI and technological innovations. His work focuses on advancing web and mobile solutions, contributing significantly to corporate learning and development by combining global insights with technology to redefine leadership and the future of work.

More Content by Romero Habib
Previous Resource
4 Ways to Enrich Leadership Development in an AI-Enabled World
4 Ways to Enrich Leadership Development in an AI-Enabled World

Blanchard's Chief Innovation Officer, Britney Cole, shares strategies for leveraging the synergistic potent...

Next Resource
Dear Next-Gen Leaders: Use AI to Empower and Inspire
Dear Next-Gen Leaders: Use AI to Empower and Inspire

People often fear what they don’t understand or feel they can’t control says Lara Dollens.