Conversational systems can range from simple informal, bidirectional text or voice conversations such as an answer to “What time is it?” to more complex interactions such as collecting oral testimony from crime witnesses to generate a sketch of a suspect. All forms of virtual assistants, including Siri, Alexa and all of their siblings also fit here. Conversational systems do not use text/voice as the exclusive interface but enable people and machines to use multiple modalities (e.g., sight, sound, tactile, etc.) to communicate across the digital device mesh (e.g., sensors, appliances, IoT systems).
As one example, a Chatbot (also known as a talkbot, chatterbot, Bot, chatterbox, Artificial Conversational Entity) is a computer program which conducts a conversation via auditory or textual methods. Such programs are often designed to convincingly simulate how a human would behave as a conversational partner, thereby passing the Turing test. Chatterbots are typically used in dialog systems for various practical purposes including customer service or information acquisition. Some chatterbots use sophisticated natural language processing systems, but many simpler systems scan for keywords within the input, then pull a reply with the most matching keywords, or the most similar wording pattern, from a database.
Opportunity Statement:
This brainy breed of conversational algorithms will leverage the power of cloud-based AI to engage and delight humans with unprecedented conversational smarts. Imagine the possibilities for everyday learners as well as special learners.
Sources:
Gartner Top Ten Technology Trends 2017
The chatbot as a tool is quickly becoming as ubiquitous in educational technology as the search bar. In my professional context, the chatbot represents a major challenge. On one hand, they are incredibly simple to implement, and have immense amounts of supporting infrastructure and clever solutions constantly being added to them. On the other, finding a way to leverage them in a way that demonstrably adds value to learners is a complex and difficult process.
For example, I have been contributing to a project looking to apply different chatbots to play different roles in a learning module – allowing for role play, but also aiming to facilitate Vygotsky’s “more knowledgeable other”. It has been no easy task.
I think it is safe to say that chatbots are not going away, and are exponentially increasing in capability – demonstrating incredible potential. Where this potential will be realized remains unclear though, as is their potential positive or negative impacts on the learning process.
At my university, we launched a new strategic plan earlier this year that requires all undergraduate and graduate courses to incorporate a chatbot. To support this initiative, we’ve already integrated the chatbot into our Learning Management System (LMS).
So far, the chatbot has proven especially useful for managing large classes. One of the features faculty members appreciate most is its ability to handle repetitive logistical questions—such as exam dates, assignment deadlines, and submission requirements. Instead of students having to search through the LMS, the chatbot provides quick, consistent answers, saving instructors a significant amount of time.
In addition, faculty can upload key course content and important concepts to the chatbot’s knowledge base. This enables the chatbot to assist students with previewing and reviewing course material at any time, making it a helpful companion throughout the learning process.
Overall, feedback has been largely positive. While some chatbots still feel somewhat mechanical and aren’t a substitute for thoughtful teacher guidance, they are proving to be valuable teaching assistants—especially for streamlining communication and supporting student engagement.
I wasn’t sure what the difference was between ChatGPT and a chatbot, so I ended up asking ChatGPT to help clarify my own understanding. Here is the key part of what Chat GPT said: “Traditional chatbots often rely on pre-programmed rules and scripts, while ChatGPT leverages large language models (LLMs) trained on vast datasets to generate more human-like and adaptable responses”. Essentially, the difference is that LLMs train from much larger dataset and they are are capable of learning from interactions with users. Given this, in my assessment of this emerging technology, I included LLMs such as ChatGPT under this category.
Within my classroom, my students don’t have much opportunity to interact with chatbots. However, with my friends’ children who range from 5 to 10 years old, I have seen how chatbots such as Alexa and Siri are so integrated in their daily lives. If there is a fact they want to learn, they simply go “Hey, Alexa, what’s______?” With the students that I tutor, though I discourage them from using chatbots for creating original content, asking ChatGPT to help them clarify their own understanding or come up with practice quiz questions for them to study for a test is an essential part of their homework routine now.
It can be beneficial to their learning that students now have what is basically a built-in AI tutor. Yet, it is scary that the information compiled and regenerated by LLMs like ChatGPT are often received and then spread without a critical eye. Despite whatever misgivings I have about AI chatbots, I think there is no denying that they will feature prominently in our educational landscape moving forward.
I am torn about this technology perhaps more than any other, which is why I think it might be the most important one for me. Chatbots are already affecting education in significant ways. Teaching teenage students in an online setting has made it especially challenging to design learning opportunities where cognitive tasks cannot be easily side-stepped by students using these tools. Although, when used thoughtfully, students use chatbots to get ‘unstuck’ or to learn about something more deeply with great affect.
While I have not seen this in practice, there is also potential with these tools to go beyond academic learning and help students practice social interactions in a low-risk environment. This could especially benefit neurodiverse students. However, there are significant safety concerns around young people becoming overly reliant upon the technology and forming unhealthy relationships with chatbots which may cause themselves or others harm. Chatbots offer tremendous beneficial opportunities to be realized in education; yet it remains unclear how to avoid the harms.