Photos are manually chosen by me. Ideally I would automate it somehow by sourcing a feed, but not every image is suitable for captioning. Some might just not be

Photos are manually chosen by me. Ideally I would automate it somehow by sourcing a feed, but not every image is suitable for captioning. Some might just not be that interesting, while others might depict things we don't want to joke about. I haven't figured out the right approach yet, so I'm open to ideas. As for battles between human caption makers and AI: I'm not sure that would be a fair battle. As far as I know AI isn't advanced enough yet to understand comedy.