Forum Discussion
Week 2 - Group 1 Discussion
https://share.articulate.com/NTslzxhG0RFsmyCfb5VrA
- 3 changes I made - 1) limited multiple choice or multiple response question-type responses to only 3 responses, 2) Eliminated on-screen text for any audio elements being used, transcripts are automatically available, 3) Attempted to space out personnel in images to make it more realistic (though I used 3D AI-generated images)
- Most valuable AI feature or capability - Image generation was very handy despite the challenges to getting the prompt just right to get what I was looking for.
- Challenges - Image generation seemed to be the toughest at times, but it really wasn't that bad.
Did you find any patterns in what prompts seemed to create better images? We're always looking for what works best!
Hi Mary!
I can't say there were any real patterns per se, more along the lines of telling it what I wanted and what I didn't want.
The thing that bugs me most with "stock photo-like" AI-generated images is that it'll drop 5-7 people all in a small space, facing "the camera", with a myriad of different technologies in front of them, with a few of them making some form of hand gesture. If it were a real situation, there would be HR violations in there somewhere. So, I often choose to reduce the number of people and try to spread them out a little to provide personal space, and that seems to work. YMMV
