New Content: Working with AI Episode

[douglowe]

Following on from feedback in #476 - we believe that we should consider including an episode on working with AI, as this will be an increasingly important topic.

Our suggestion would be that this should be a high-level discussion, at the end of the workshop (perhaps even the last episode of the course). It should stress that following best practice is even more important when working with AI, and lead into an open discussion within the groups on the benefits and drawbacks of working with AI tooling.

If we want to have recommendations for learners in this episode, we think these probably should be (non-definitive list):

• Use AI for 'rubber-ducking', questioning error messages, etc
• Be very careful of AI within IDE's - don't just blindly accept all the suggested code blocks that will be thrown up by it
• Be aware that AI can halluciate, check library suggestions!
• Consider Copyright and Licence requirements! What implications do your copyright / licence restrictions have on the use of AI. Consider the licence that you've signed for using any AI tools, and if that conflicts with your organisations licence.

[matt-graham]

Some possibly relevant discussion on content on AI tools on Carpentries blog

[bielsnohr]

Another link from a maintainer meeting is what the eScience center has done here: https://github.com/olgaminaeva/gen-ai-coding/tree/main (thanks @JaroCamphuijsen )

[bielsnohr]

From maintainer's meeting on 2025-09-23, a few further bits of information were raised:

• @amangoel185 is going to consult with what will be done on Building Better Software course
• we should keep a close eye on what gets developed for the new lessons coming into the Carpentries led by Toby
• we won't want a "technical" episode on this; rather, it will be mostly signposting and a bit of general advice