Recently I needed to provide a description of a process that we’re going to implement for an upcoming conference, but the process we’d developed in our meetings was written as a list i.e. short, incomplete sentences in bullet point format. We wanted to submit a narrative description so I took that outline and gave it to Claude as an attached text file. All of the relevant information was included in the attachment; it was just incomplete and poorly written.
This is the prompt I used: “Please take the attached file and write me a 1000 word overview describing the process in more detail. I would like you to include a rationale for the decisions made, and also some clear descriptions about how this approach supports a more inclusive way to design conference events. I want you to emphasise the value in getting input from participants, before, during, and after the conference. I also want you to explain why it’s important to pay so much attention to the opinions of so-called experts, especially in fields that are so new that it’s unlikely that any of us really has any expertise.”
This is a perfect example of using generative AI to save time. I could have written the overview myself. The team I was working with had done the work to prepare the outline so I would know if Claude introduced errors into the description. Basically, Claude wrote the first draft of what I would have written if I had taken the time to do so. I ended up spending a couple of minutes editing the output, for a total of 5 minutes to get it all done. I estimate that it would have taken me at least half an hour to convert the process into a narrative.
We need more academics using more AI for more tasks.