Imagine a world where personalised education is at your fingertips. AI chatbots are making this a reality, but are they the future of learning?
What are AI chatbots?
AI chatbots are essentially online assistants which answer questions and converse with users. Many people will have tried ChatGPT, but chatbots have been around for years, like the customer service bots found on websites. Today, Large Language Models, which power ChatGPT and other generative AI, have developed to a level where they can provide nuanced, context-aware responses. This makes them highly effective for supporting personalised learning experiences and facilitating learner engagement.
How can they be used in education?
AI chatbots are not yet commonly used in education, though they’re becoming more useful and accessible as the technology develops. Here’s a closer look at some of their applications.
Teacher or facilitator
AI chatbots can facilitate a dialogue-based learning approach, similar to that of the Socratic method. This teaching approach involves guiding learners through questions and dialogue, encouraging critical thinking and deeper understanding.
AI chatbots can help learners connect new information with their existing knowledge, evaluate their understanding of a topic and reflect on their learning. This isn't about substituting human teachers or classroom discussions but rather supplementing and extending the typical classroom experience.
Tutor
AI chatbots also excel as personalised tutors. They can adapt to the specific needs or level of individuals, offering tailored advice and guiding learners through problem-solving processes. For example, if the learner can’t solve a Maths problem, it can give them step-by-step support in solving the problem using the appropriate method. AI tutors are particularly useful for learners during independent study where a teacher isn’t always available, such as large classes, homework or remote communities.
Performance support
In corporate training, AI chatbots can enhance practical learning and performance support. They're especially useful in creating realistic training scenarios, which can be more authentic than traditional e-learning modules or roleplays. For example, a chatbot can simulate a customer in a sales training exercise, providing a safe space for learners to practise realistic interactions.
When AI talking-head videos are linked with Large Language Models, these scenarios become even more lifelike, allowing learners to interact with AI avatars that mimic real customers. The feedback provided by AI is immediate, specific and unbiased by personal relationships or workplace dynamics, offering a unique advantage over traditional feedback mechanisms.
And much more…
Beyond serving as teachers, tutors or performance support tools, AI chatbots have a multitude of other educational applications, including:
Providing general learner support, such as administrative assistance or subject-related queries.
Assisting in formative and summative assessments.
Offering real-time workplace learning support, guiding employees to relevant resources, contacts or policies.
Facilitating spaced practice with scheduled activities to enhance learning retention and reinforce knowledge.
How do AI chatbots help us learn?
To get the best out of AI chatbots for your learners, it's important to understand how they affect various aspects of the learning process.
Feedback
Immediate and appropriate feedback is fundamental for learning, and chatbots excel in this area. They provide real-time responses to learners' inputs, enabling a continuous feedback loop. This immediacy helps in reinforcing concepts, correcting misconceptions and guiding learners through the learning process more effectively.
Motivation
The immediacy and interactivity of chatbots keep learners active and in control, fostering a sense of autonomy and engagement. This contrasts with traditional learning, where learners often have little input or control over the experience. Personalised chatbots cater to individual needs, backgrounds and attitudes, making the experience more relevant to the learner.
Personalisation
Personalisation is one of the key strengths of AI chatbots in learning. While teachers may find it challenging to cater to each learner's unique needs in a group setting, chatbots can offer tailored support anytime, anywhere. They're available 24/7, aiding with homework, revision or independent study. Conversations with the chatbot can be analysed to inform both the educator and the AI, further enhancing future learning interactions.
Desirable difficulty
Desirable difficulty enhances learning, as it challenges learners just enough, reducing boredom, confusion and frustration. Chatbots can guide learners towards answers instead of simply providing them, mirroring a teacher's scaffolding approach. If a learner struggles, the chatbot can offer step-by-step guidance, adapting to their level of understanding.
Attention
AI chatbots generally present information in manageable, conversational chunks, tailored to the learner's level and language ability. This makes it easier for learners to focus on and engage with key information, which increases the likelihood that learning is remembered. The user experience with chatbots is typically straightforward and familiar, making it easier to focus on the learning. A seamless user experience is crucial, as a clunky or confusing interface can distract learners, reducing their motivation and potentially causing them to disengage.
Accessibility
As technology advances, more chatbots are multimodal, presenting information through audio or visuals, as well as text. This can support learners with diverse needs. Chatbots are also available in multiple languages allowing multlingual support when needed, making learning content more accessible to a global audience. Additionally, chatbots' scalability means they can be deployed widely and cost-effectively, potentially democratising access to education.
Spaced repetition
Chatbots can implement spaced repetition, known for enhancing long-term retention of information. By presenting material at strategically timed intervals, they ensure that learners review and reinforce their knowledge making it easier to recall later. Chatbots can automate this process, delivering content at optimal times, making the learning process more effective.
What are the downsides to using AI chatbots?
While AI chatbots offer numerous benefits for the learner, it's important to acknowledge their limitations and potential downsides.
Cognitive effort
We’ve seen above that chatbots can support critical thinking and understanding, but if not set up appropriately, they could undermine the cognitive effort required in the learning process. Effective learning often involves actively grappling with challenges and engaging in deep thought, known as ‘cognitive effort’. If information is too readily available, chatbots could reduce the need for critical thinking, independent research and summarisation skills.
Social learning
Chatbots typically facilitate one-to-one interactions, which can limit opportunities for social learning. Social learning theories emphasise the importance of learning through observation, imitation and modelling of others. Chatbots also may not the best approach for fostering interpersonal, communication or collaboration skills. While chatbots can use the Socratic method to promote critical thinking, they cannot currently reproduce the dynamics of group discussions and debates.
Human connection
The role of human educators extends beyond imparting knowledge; they inspire, understand and connect emotionally with learners. Human teachers play a critical role in building relationships and can understand the nuanced emotional needs of their students. Chatbots, cannot fully replicate these aspects of education, making them a supplement, rather than a replacement, for humans in learning environments.
Limited subject depth
While AI chatbots are adept at covering a wide range of topics, their depth of knowledge in specialised subjects can be limited. In Higher Education, advanced studies and some corporate environments where in-depth expertise and nuanced understanding are required, chatbots might not provide the level of insight or complexity needed.
Incorrect or harmful information
AI chatbots can sometimes 'hallucinate' or provide incorrect information, which learners might accept as fact. In some instances, they might even offer harmful advice or content. For example, a chatbot used by an eating disorder hotline was found to provide potentially dangerous advice. This raises concerns about the reliability of information provided by chatbots and underscores the need for careful training and supervision, especially when they are used in sensitive or critical learning areas.
Bias
Chatbots may exhibit biases present in the algorithms and data used to train them. A lack of diversity in the training data, or even in the engineers making the algorithms, can reinforce stereotypes and exclude certain viewpoints. For example, image generators may generate more images of caucasians or create characters that represent stereotypes. For learners, this bias can lead to a narrow, non-inclusive learning experience that doesn't reflect the diverse realities of the world.
Privacy Concerns
The use of chatbots, especially those requiring personal data to personalise learning experiences, raises privacy concerns. Ensuring the confidentiality and security of learner data is crucial, and any breach could have serious implications for both learners and educational institutions.
Dependence on technology
Heavy reliance on chatbots for learning can lead to an overdependence on technology, potentially reducing opportunities for learners to develop problem-solving skills and resilience in the absence of digital assistance.
So, are AI chatbots the future of learning?
AI chatbots can give a whole new dimension to the learning experience, offering engaging, personalised and interactive one-to-one support whenever and wherever the learner needs it. While AI chatbots represent a significant step forward in educational technology, they are more suitable in particular situations, such as tutoring, rather than a one-size-fits-all solution.
Want to learn more about how AI can be used for learning design? Don't forget to check out my other posts and subscribe using the button in the menu or footer.
Comments