Julia Slay

Last week I facilitated a workshop where a guest speaker talked about how AI would change the world of work for charity fundraisers. It was thought provoking stuff! This week, the BBC published a story on how AI might transform the legal profession, and so I decided to spend a week experimenting with it in my daily work to try and get a sense for how it might influence my work as a facilitator and researcher. Here is a brief roundup of how it went – and my conclusions so far! 

Monday 

• AM: working up a facilitation training outline for an in-house client. I want to craft something bespoke for this potential client so I am not optimistic about how well AI can help, but I insert some text from previous outlines and it generates a workable enough structure for the training. It’s not quite what I’m after though, so goes in the ‘virtual’ bin. 

• AM: briefing call with a client about a new piece of desk research. I half heartedly ask Chat GPT what questions they would ask a new client during a meeting, and am surprised by the reminders it gives me on useful things to cover. That said, I’m not sure it’s much better than something a simple Google search would have done.

• PM: delivering a workshop to develop an organisation’s impact framework and on the tube to their offices I ask chat GPT to generate some examples of outcome indicators for the framework. It gives some helpful high level areas of indicators, but not the level of detail or specificity I’m after. 

• PM: post workshop I have a lot of notes that I’ve captured which I want to summarise nicely and share back with the client. Bingo! This is the most helpful thing it’s done all day. It’s not perfect, but it’s 70% of the way there, and makes my job more of an edit piece. I still have to spend some time on it, but it has removed the mental load of the initial write up. 

Tuesday – my partner upgrades me to his Premium ChatCPT account! 

• AM: I am reviewing some strategy documentation for a new project and want to create a summary on hand for future reference. I copy and paste the text into ChatGPT and ask it to create a 10 point summary. Bingo! This is helpful, and is something I can imagine coming back to these in future. 

• PM: writing a bid. I’ve just been sent a tender from a contact for a piece of work which I’d love to do, but the deadline is the next morning. I don’t think I really have time to write a good response, but I’m curious about how well AI will fare on this front. I copy the tender text and put it into Chat GPT and ask it to generate a 2 page bid. It does a good job of generating 40% of what I would usually put into a bid to structure it and capture key elements of the proposal, but it definitely isn’t good enough to submit without more work. 

Wednesday 

• AM: I am getting ready to submit a bid I have been working on for a while, and just before I press ‘’submit’’ I copy the text into ChatGPT and ask it to critique it from the perspective of a grants assessor and identify weak areas. It comes up with a fairly generic list of things to consider, which don’t feel all that helpful! 

• PM: I am reviewing a draft learning framework and I ask Chat GPT to identify any weaknesses or gaps in the current version. It identifies a couple of really interesting areas that I haven’t spotted, and feels like getting a helpful pair of extra eyes on my analysis. 

Thursday 

• AM: I am starting to pull together the learning pack for some facilitation training I am running and I ask Chat GPT to recommend some resources I can include and it generates a list of books, but I am looking for more practical mateirals, videos, tools and blogs which seem out of its capacity right now. 

• PM: switch off the laptop and head to Paddington for a train out of London, and a weekend off grid!