Can an AI tool build a Flow matching best practice standards? If this sentence intrigues you, then this session will be worth checking out. In it we will ask OpenAI's ChatGPT tool to put together instructions for building a Flow based on a use case. We'll check out the results together and see how it compares to current Flow best practices. You might find the results a little surprising.

ChatGPT has been in a lot of conversation in and around the Trailblazer Community over the last number of weeks. As someone who has a major focus on Flow, I recently experimented with asking ChatGPT to provide instructions on how to put together a Flow based on a requirement. This brought about some mixed results and some content that will be coming to my blog very soon. This is what led to the idea for this session. The main idea for this session will be to select from a selection of use case requirements, submit them to ChatGPT and compare the results to Flow. Depending on the setup I'm thinking of having this be interactive and have the attendees select the Flow use cases submitted to ChatGPT. The basic makeup will be to submit the use case to ChatGPT, compare that to best practices in Flow Builder and review where the gaps are. Depending on time, I'm thinking that it could be good to even demo the Flows and show them in the backend and maybe running. This session will prove insightful to the attendees as it will give them a flavour on the accuracy of an AI tool such as ChatGPT. It will prove intriguing to those interested in AI and declarative automation in Salesforce using Flow. It should also prove to be a fun session.