Table of Contents
Recommendation
Learn more about the potential and challenges that lie ahead for adaptive learning and its intersection with AI. In this episode of The Learning Hack, host John Helmer delves into adaptive learning systems and their connection to AI with the Chief Evangelist of adaptive vendor Obrizum, Markus Bernhardt. Helmer and Bernhardt explore whether adaptive learning’s future looks promising or apocalyptic in light of recent AI developments. Bernhardt provides some reassurance while also highlighting crucial concerns related to AI’s use in corporate settings – including its accuracy and reliability.
Take-Aways
- Adaptive AI can personalize learning at scale.
- AI technology will enable real-time performance support.
- Generative and non-generative AI serve different functions but require similar safeguards.
Summary
Adaptive AI can personalize learning at scale.
Adaptive and generative AI technologies have the potential to change corporate learning dramatically. AI systems can construct instructional maps and deliver adaptive workbooks to learners by analyzing and connecting related topics. It can act as an individual human tutor might, providing tailored exercises, content and practice questions based on learners’ strengths, weaknesses and progress while also considering their self-reported confidence levels.
“All of the things a tutor would do, now the AI is doing by saying, ‘I’ll ask you some questions, I’ll show you some content.’ It adapts in the delivery of those learning and practice questions in line with the strengths and the weaknesses and the rate of progress of the learner.” (Markus Bernhardt)
In this way, the technology works similarly to satellite navigation, pinpointing where learners are and where they want to go. This unique approach can maximize learners’ competence and confidence, ensuring they can apply their knowledge well in real-world situations.
AI technology will enable real-time performance support.
When it comes to corporate learning, the ideal is to achieve a level of learning personalization akin to what a member of a professional sports team enjoys. Professional athletes have fitness coaches, nutritionists and technical trainers who create highly personalized plans for each player that also take the team’s needs as a whole into account. These experts analyze performance on a play-by-play basis, exploring what’s working and what isn’t in real time.
Budget limitations hinder the employment of one-on-one super specialists for each individual within most corporate settings. This is where integration of AI-driven adaptive learning technologies can help in significant ways – serving up in-the-moment, tailored performance support, as well as more formal learning.
Generative and non-generative AI serve different functions but require similar safeguards.
While generative AI tools like ChatGPT can be valuable for certain tasks, they are only suitable for some learning contexts. For instance, using ChatGPT to translate podcasts can be a brilliant approach for immersive language learning. However, relying solely on GPT for complex tasks like generating medical surgery instructions would not be advisable.
Non-generative AI does not search the internet or draw on information outside its predetermined context. It uses preselected information – a knowledge or training set, for example – and then maps out learning paths based on learners’ needs. It delivers personalized learning but keeps it within a specific context. This ensures the information available to learners is accurate.
“[L&D professionals] need to be comfortable with conversations around…risks and what the AI has access to, how it works, where it saves things, how it works with feedback, and how this information is kept safe and ring-fenced and [is] utilized.” (Markus Bernhardt)
However, L&D professionals must remain aware that the data used to train AI may contain biases even when dealing with non-generative AI. They must also address user privacy concerns. They should ensure that data produced from training sessions remains within the organization. By openly addressing issues such as data privacy, companies can foster employees’ trust and confidence in AI-powered learning solutions.
About the Podcast
Author and host John Helmer is a writer and show-runner of The Learning Hack podcast and Markus Bernhardt is the Chief Evangelist of adaptive vendor Obrizum.