Using acceptance criteria to QA AI-generated content
How often have you generated something on ChatGPT that’s completely unacceptable?
This article was co-authored with Annelie Tinworth, Lead UX Content Designer at Volvo Cars, localisation legend, and word nerd extraordinaire.
*Ahem*
Be right back, too busy revelling in the sweet magic of a brand’s unique value proposition to write right now.
But seriously, while we spend a lot of time creating frameworks and practising techniques to improve output, it’s still pretty dodgy a lot of the time. Sometimes, it’s just easier to continue the writing process outside of the tool. Still, we all get writer’s block every once in a while, or we’re tasked with writing something complex before the end of the day on a Friday afternoon. How can we keep leveraging ChatGPT to brainstorm?
Enter acceptance criteria. Let’s take a look at how these can help you judge the quality of your generated content with respect to what it needs to communicate, and how you can continue refining your output to get something more usable.