Blog

Learning: how AI can help us get through the crisis

While many companies are wondering what their activity will look like in the future, some fields have already started reinventing themselves to adapt to the crisis and minimize its impacts. Where does learning and development stand?​

One major challenge: learning better and faster

Once the pandemic is over, companies will have changed a lot and they will need to adapt to avoid (or limit) the crisis. In the field of learning and development, two challenges will have to be met by companies: enhancing the knowledge management in contexts of reorganization and continuing to develop their employees but in a more optimized way. Artificial Intelligence represents a real opportunity.

Once the pandemic is over, companies will have changed a lot and they will need to adapt to avoid (or limit) the crisis. In the field of learning and development, two challenges will have to be met by companies: enhancing the knowledge management in contexts of reorganization and continuing to develop their employees but in a more optimized way. Artificial Intelligence represents a real opportunity.

For years, digital has enabled to optimize costs by expanding the reach of learning: tens of thousands of employees can follow the exact same learning path at different times and in different locations. But this expansion sometimes lowers the learning and teaching quality. On a large scale, all learners are addressed in the same way, no matter their job, skills proficiency, seniority, etc. It means that an employee with a fifteen-year experience sometimes ends up spending hours in training – to acquire the skills required for his job – whereas ten minutes would have been sufficient to acquire the ONE missing concept.

What data to identify learning needs?

The question to ask is the following: how can we keep expanding the reach of learning while adapting it to the needs of each person, as well as the performance needs of the company? The combination of the words expansion and individualization often rimes with Artificial Intelligence. But Artificial Intelligence is nothing without data. Two types of data are essential to identify the upskilling needs of a person or a group of persons:

• The knowledge and skills objectives according to the company’s current jobs (and future jobs when they can be predicted). Usually in the form of repositories, they define the requirements of an employee based on his position or a future position expected by himself or the company.
• The current proficiency of an employee on the knowledge and skills he is supposed to have. When comparing the employee’s proficiency with his objectives, you get his skills gap of learning need.

For instance, a technician who works with customers has to be an expert in the regulatory knowledge and technical procedures related to his scope of action, and he must have some customer relationship skills. If his proficiency on customer relationship is assessed as sufficient, then he will not need to develop more on this topic; but if the assessment shows a medium proficiency on the technical procedures, then he will need to learn to achieve the expected level of expertise. Today, there are many methods to assess this proficiency: traditional quiz, self-assessment, assessment interview, 360-degree feedback, etc.

Concrete applications of Artificial Intelligence

As you probably understood, the main challenge is to rationalize learning actions. In this way, Artificial Intelligence allows to address to major challenges in learning and development.

First, AI automates the assessment of employees’ knowledge and skills to identify which one of them they need to develop. We talk about adaptive positioning test: unlike a traditional assessment, it defines the employee’s level of proficiency on the knowledge he is supposed to have by asking the most adapted questions according to his profile. Therefore, the algorithm automatically selects different questions to assess a technician and a sales representative on customer relationship, since their job requirements are different and their proficiency is probably different as well. In summary, to assess different profiles on a same topic, the AI will choose the right questions to ask to provide an individualized and optimized (shorter) quiz.

Once the skills gap is identified, the employee’s next step is to follow the right learning path: a difficult task when the course catalog is composed of thousands or even tens of thousands of learning modules. Artificial Intelligence also supports this, in the form of a recommendation engine. When it is identified that the employee needs to develop himself in one particular skill, the AI is able to react immediately by suggesting to the employee the available learning module(s) within the catalog which will optimize the upskilling process.

Investing in learning to support the transformation

If we take one step further, AI will be able to predict the format (a one-day training? an e-learning spread over one week? five micro-learning modules?) which will have the better impact on the employee’s progress, based on the past experiences of other employees with similar profiles. Therefore, the recommendation engine builds a tailored learning plan for each employee to focus on their individual needs and reduce the time needed to acquire the key skills.

As a conclusion, in a context of crisis and deep transformation for companies, it has become crucial to encourage employee upskilling to develop the digital culture, enhance the knowledge management and improve the agility of organizations, among other things. Of course, this context also impels companies to rationalize the resources and time dedicated to learning and development: Artificial Intelligence is one answer to this challenge as it automates some learning processes and individualizes learning paths.