How does AI benefit the Training and Development industry, and how will AI impact multiple stakeholders?
Artificial Intelligence (AI) has been a fascinating area of learning and application for decades now, starting in the 1950s. Researchers and industry experts have tried applications of AI in various industries such as manufacturing, automotive, retail, financial services, healthcare and others. AI can automate mundane tasks so that humans can engage and take on higher-value, qualitative work. The fundamental question posed by the visionary Alan Turing — Can machines think? — is coming closer to reality as advances in AI continue apace.
We have started moving AI applications built with narrow intelligence toward more general intelligence and next toward super intelligence. As shown on the timeline in Figure 1, we are now deep into the emergence and delivery of personal consumer AI-based solutions such as Alexa, Siri, and Google Now, and in particular of more complex, higher-value Enterprise AI and cognitive systems such as IPsoft’s Amelia.
Being part of the Education, Learning, Training and Assessment industry for two decades, I often ponder how does AI benefit the Training and Development domain, and how will AI change it over time. I’m not specifically referring to training an AI system itself with the necessary data and business expertise for a particular company. I’m focused on what value AI can deliver in training humans in a variety of programmatic roles and functions. I outline below some of my thoughts on how AI can be integrated into the different areas of Training and Development in the near and long term.
Benefits Across Training Ecosystems
The stakeholders of these AI-powered training benefits across an organization fall primarily into three categories: end users (learners), the training function and the organization’s management team. Each of these categories have their own unique training needs and goals, but share some common potential benefits for engaging AI in training — namely, saved time and more efficient use of resources. As for functional areas, examples of potential AI impact are summarized in Figure 2, spanning content development, training delivery, training operations and training outcomes/results. Specific examples for each are below.
Training Need Analysis (TNA) and Content Development/Instruction Design
Learning Needs Assessment: AI can gather an understanding of learner needs through Learning Management System (LMS) analytics data, and make suggestions for additional training or how existing programs can be modified for greater usage and applicability.
AI-assisted content development based on facts/data learning: Content development tools (e.g., instruction design [ID] tools such as like Articulate, Captivate, Camtasia, etc.) can have integrated AI modules/plugins that assist (by learning) an ID developer with creating content. The AI can suggest templates, storyboards, suitable graphic elements, text arrangements and so on. This can lead to stronger content creation and improved production thanks to an AI-based, data-driven approach to learn what content elements are most effective.
AI-assisted Text-to-Voice (of any language): AI’s ability to learn and understand language can enable faster content creation through the conversion of text-based material to voiceovers in Web-based content development.
Training Content Delivery
Trained Virtual Instructor/ Co-Instructor: Broadly speaking, AI is often viewed as replacing human work and roles across businesses and industries. In truth, AI will more often be used as an augmentation to existing employees, not a replacement, with the establishment of collaborative AI/human hybrid teams. The same holds true for training delivery. While AI can take the lead in small duration courses related to compliance or technology overviews, AI can work with human colleagues on more complex training material such as certification, with AI automating certain training elements and humans providing unique input when necessary.
Training Operations (Day-to-Day Management Activities)
Learner-based services: AI — through Robotic Process Automation bots as well as modules and plugins for a LMS — can engage with users for automatic class scheduling, roster creation, communication, notifications and class closure/administrative activities based on learner data.
Guided learning and FAQs: With AI acting as a whisper agent integrated with a LMS and accessible via a chat interface, learners could receive information on courses, course requirements and timescales, as well as guided training assistance within specific courses. This kind of AI deployment is quickly becoming commonplace in other industries; Amelia already acts as a whisper agent for human call center workers in insurance and banking.
Course feedback and improvement: Collecting learner feedback is vital in order to enable a continuous improvement cycle for training and development efforts. AI can automate feedback/survey collection on courses and training through a chat-based interface, with questions specified to the subjects covered. In addition, AI can provide insights through analytics, identifying the most popular course or training elements and potential areas of improvement.
These use cases represent some of the potential near-term deployment for AI within different areas of Training and Development. Longer-term, additional use cases and benefits will emerge as AI gains traction within various corporate training groups as well as universities. We may still be a few years away from answering whether AI can think, but it’s clear that AI can help people learn in faster and more efficient ways in the here and now.