For Day 2 of The 12 Days of AI, the 20 minute task is to use an AI tool called Claude.
To complete today’s task I need to upload a Word document or PDF to Claude and ask it to summarise the text.
Cool.
Asking Claude to summarise my 12,000 word research paper
Using the affordances of Claude, I was able to summarise my 12,000 word final research paper ‘Our course could be your life(style): Tales of MOOC revisiting learners 2015-2020’.
It’s worth noting that there was an additional task for Day 2 to summarise everyday tasks such as “contents or the individual contribution of someone in a online seminar or supervision meeting”, which is pretty cool, but given I’m away from work I didn’t have a transcript readily available – I’ll revisit this on my return to work in 2024, okay?
With the prompt “Summarise the research paper in no more than 200 words, including a title and sub-heading”, Claude produced the following summary of my final research paper:
Revisiting the same MOOC: Exploring motivations of repeat learners
Abstract
MOOCs allow learners to freely re-enroll in courses. This study explores what motivates learners to repeatedly revisit the same mindfulness skills MOOC.
Methodology
Learner comments over 14 runs of a mindfulness MOOC were thematically analyzed to identify motivations for re-enrollment. Case studies of 4 committed revisiting learners mapped engagement over time.
Key Findings
Analysis identified core motivations like community, personal practice, content updates, and emotional benefits. Two revisiting patterns emerged: consecutive “sequencers” and sporadic “syncopators”. Both valued social connections and reminding themselves to continue mindfulness practice. Revisiting provided ongoing motivation, skill reinforcement and positive emotional impacts for these learners.
Conclusion
Beyond initial interest in content, learners revisit MOOCs for community, updated materials, and lifelong skill development. Supporting committed learners undertaking continuous mastery journeys should be a design priority. Further research could more directly survey revisiting learner motivations.
Reflection on the Day 2 task
Like others participating in the Day 2 task, my thoughts and reflections on the experience of using Claude to summarise text were similar.
Did I find the task useful? Sure, I did. Learning how to use an AI tool to kick-start the summarising process is a good thing.
Does this use of AI undermine the development of the key writing skill of summarising and paraphrasing text? Perhaps the production aspect of the skill, but not so much the thinking – what’s produced with the tool still needs to be reviewed for accuracy and meaningfulness.
Should Claude be used by and and promoted to students as tool for them to use? Sure, learning how to use the tool and understanding its strengths and weaknesses and how it can be used in the academic and professional life is a good thing – using the medium to critique the medium.
What about privacy concerns with using Claude or ChatGPT? Sure, there’s always an underlying concern that my data is most likely providing further training for the conversational model, but that could be considered the price to be paid to make use of the service – you cannot get something for nothing, right?
What about the quality of Claude’s summaries – were they any good? I used Claude to summarise my 12,000 word research paper, and given that I wrote it and have in-depth understanding and first-hand awareness of the content, I felt Claude’s summary was of high quality. I’d be a little more cautious about double-checking and quality assurance if I was less familiar with the content or didn’t have first-hand experience of the source material, which you’d probably do anyway as part of process even if you weren’t using AI tools.
How else could Claude be useful in a professional practice? Like any tool, Claude could be used as part of a larger production workflow and in combination with other tools like ChatGPT.
And as one other learner noted in Day 2, Claude can be used in combination with another AI tool like Chat GPT to produce a blog post. From the initial Claude-generated summary, the prompt and summary could also be used to then produce a script for a podcast or video that then is passed through to another AI tool to create the audio, music bed, video and related visuals – it really depends on what your goal is and you mix and match the tools as needed to get the job done.
Optional readings
Oh, yeah. There’s some recommended optional reading that’s probably worth reading, too – it’s Assessing AI-based Summarizing and Paraphrasing Tools for a CLIL Intercultural Communication Academic Writing Class by Barry Kavanagh (Associate Professor at Tohoku University, Japan).
