Scroll down

How artificial intelligence has entered the Learning & Development market

Margaux Daza |

The use of artificial intelligence (AI) in Learning & Development (L&D) is a matter of debate as some do not see the potential of “Big Data” in this field: the data is abundant but sometimes hard to collect and rarely qualified. Then, why not talk about “Smart Data” instead?

Smart Data is an approach which consists in computing in real time exclusively qualified data and allows L&D departments to gain in efficiency and provide a training which actually meets employees’ individual needs.

Learning already has experience with AI

One of the reasons why AI has entered the world of Learning is because companies have plenty of data to be analyzed: LMSs and other learning platforms have been implemented for years which has allowed to track employees’ learning activities. Those learning activities represent highly qualified data which we can call Smart Data. The added value of AI is to automatize some tasks which would be time‑consuming (or even impossible) for instructors, managers and HR departments.

Machine learning algorithms are used to compute this Smart Data. The algorithms are similar to those used in the fields of marketing and finance for instance. But it is important to take account that AI is far from replacing L&D departments, instead of that it assists them to meet the skills needs of companies and employees.

Read about the origins of AI in the article “Introduction to Artificial Intelligence

Engaging the employee in learning

Just like platforms such as Netflix or Amazon which aim to define your user profile to suggest adapted content, the use of AI in learning aims to define your learner profile to recommend a personalized learning path. And there is a variety of challenges to address, for instance: one-size-fits-all approaches in learning sometimes downgrade its quality and relevance; designing learning plans for employees is time‑consuming for some managers; some e-learning courses, designed to reduce costs, are too long and cannot engage the learner until the end.

The analysis of data allows to understand the learning behavior and needs of an employee in terms of skills, in order to provide a learner-centric experience and to encourage the fact that employees learn on their own. Here are some examples of functionalities which can be implemented from this analysis:

  • Individualized learning plans according to each employee’s level of proficiency and the company’s current and future needs in skills.
  • Personalized e-learning courses to optimize the employee’s path based on his/her starting level and progress.
  • Smart assistants to support the employee on the long term with personalized reminders on skills acquired during his/her career and likely to be forgotten.
  • Recommendations of learning material designed by the company or not, adapted to the employee’s interests, based on his/her past experiences and on the experiences of other employees who are similar to him/her, to improve engagement on the learning environment.

Supporting the production of learning material

To implement a user-centric learning strategy, another requirement for engaging learning paths is to provide quality learning material while taking advantage of the different existing formats (both face-to-face and online). But designing new content and making it available in a variety of formats is time‑consuming and expensive. AI has a huge potential to automatize some part of content creation but it is still in its earliest stages.

The technologies we are talking about are NLP and NLG: Natural-Language Processing and Natural-Language Generation. The first one will allow L&D teams to save time in indexing existing content. When analyzing the learning material, the topics addressed can be found and the skills taught can be deduced: this tagging is essential for the recommendation engines mentioned above. Beside the content already available in the company’s course catalog, it would even be possible to search a video on YouTube, then automatically tag the skills addressed in it and integrate it in the catalog. As for NLG, it will allow to go one step further and automatically generate questions from a learning object: these questions can be used for assessments or spaced repetition systems. L&D teams will of course remain in charge of selecting and validating the questions.

The benefits brought by AI are to be viewed as a help to L&D departments and it is fundamental that the tools implemented by companies come with the right analytics and reporting to give visibility to the learner, the instructor, the manager and the HR department.