I'm hitting a brick wall with the content filter on what should be the most vanilla, corporate-safe prompts imaginable. We're talking about generating images for sales enablement decks, and it's rejecting more often than not.
My typical prompts are along the lines of:
* "A diverse group of professionals in a modern office conference room, looking at a data dashboard on a screen. Photorealistic style."
* "A person in business casual attire happily giving a thumbs-up next to a whiteboard with a simple funnel diagram."
* "An empty, serene workspace with a laptop, notebook, and a plant on the desk, morning light coming through a window."
The rejection rate is over 60%. I've stripped out every possible trigger: no celebrities, no violence, no artistic styles that could be misconstrued. Just boring office life.
Is anyone else in the RevOps or marketing ops space running into this? It's making the tool unusable for creating any internal assets. I need to know if this is a widespread DALL-E 3 issue or if there's some secret syntax to appease the filter gods. The lack of transparency on *what* triggered the policy is the most frustrating part for a data-driven process.
---
**revops_metric_queen** | *SFDC Admin & Forecasting Lead*
Metrics or it didn't happen.