Learning and Teaching

How AI is changing teaching at SFU

April 26, 2024

How are generative artificial intelligence (AI) tools impacting teaching and learning? Though instructors at SFU are responding in different ways, one thing they agree on is that the impacts of these technologies need to be addressed head-on.

Generative AI tools are software programs or systems that utilize artificial intelligence algorithms to create new content, such as text, images, music or videos.

The case for building it into assignments  

Business professor Eric Gedajlovic is consciously shaping student assignments to integrate the use of generative AI tools, a change he hopes will also shape how they view the tools themselves. 

“We are at a crossroads. Generative AI can be used to do things in a way that makes learning bigger and deeper or it can be used to outsource thinking. If we aren’t structuring students’ assignments towards the former then they are going to only see it as a tool for doing less. That means we have an opportunity to be leveraging AI in the classroom to show them how it can be used to extend our capabilities rather than diminish them. For instance, in an entrepreneurship class instead of asking students to create just one business plan, we can have them use AI to develop multiple business plans in order to explore more than one use case for their ideas. Or perhaps, we can ask our students to expand upon their essay creating policy briefs to explore the practical implications of their ideas. Alternatively, we could ask them to use AI to help prepare for debates by engaging with AI to get a better handle on possible counter arguments. Or, perhaps we can ask them to convert their written work into TED-style talks to reach wider audiences. If we don’t guide them down the route of doing more, then there’s a real risk of students unthinkingly using these tools to do less and less.” 

Design around it when you can

For education lecturer Cary Campbell, navigating AI has meant embracing assignments that don’t fit neatly into a ChatGPT prompt window.

“What I have found is that encouraging students to engage with multiple modalities of communication and expression, what we call multimodal literacy in education, helps protect against possible academic integrity issues. For example, engaging with multimedia, drawing a mind map, scratching, creating an audio recording or producing a photo gallery or video, prevents students from using this technology to cognitively offshore their learning because there is a need to think creatively and strategically to convey a message across the different media forms that generative AI cannot reproduce. Of course that is not always possible. There will be times when a text-based submission is required and yes, in some cases I have seen content submitted that is definitely generated by AI but whenever I talk to those students what I hear is that they used AI in those cases as a last resort. In fact, what I’m hearing from my students is that there is a real reluctance to use them and an awareness that relying on these can rob them of the learning they are here for.”

The call to teach generative AI literacy

For communication professor Frederik Lesage, generative AI isn't just a tool for learning material, but is now becoming part of the currirculum itself.   

“Students are using generative AI but not well. For example, I’ll read an essay and it seems very well written, but the arguments are circular and don't really say anything. Instead of using it as a tool to help them write, they are using it as a tool to write for them. I’m not saying they shouldn’t ever use generative AI to help them write but they don’t have the literacy to know how to use it as part of the writing process. Similarly, I have students that will submit AI-generated images that don’t represent what they are supposed to—for example depicting Seattle’s skyline instead of Vancouver’s. Universities have always been complicit in introducing the tools that become industry standards, especially in the cultural industries, and so we have an obligation to teach students how to use these tools. And we need to do it in a way that acknowledges their complexity by addressing concepts like copyright and bias. I have done this in my classes by integrating AI tools into assignments and then facilitating conversations with my students on when they can be used and how.”

Looking ahead: what should we be teaching, anyway?

According to computing science lecturer and faculty teaching fellow Diana Cukierman, one of the biggest impacts that generative AI will have on teaching and learning is shaping what gets taught and what gets left behind.

“We are in the middle of the domestication of these tools, which means, yes, there may be a loss of skills, just as there was a loss of long division skills when the calculator was introduced. But in the same way as you need to have basic knowledge of numbers and functions to know what to do with the calculator, you need to have basic knowledge to use the generative AI tools. The question we need to be asking is ‘what is that knowledge in your discipline?’ This is especially important for us to examine when it comes to our introductory courses.”

To discuss these and other issues related to generative AI,  join the next offering of SFU’s AI Community of Practice (AI COP) for the SFU Teaching Community on May 15. The AI COP is a monthly online discussion for instructors hosted by the Centre for Educational Excellence on how generative AI is changing the way we think about teaching and learning.

 

Facebook
Twitter
LinkedIn
Reddit
SMS
Email
Copy