The woke nonsense isn't the source of the problem, it's a symptom of a deeper problem. The Gemeni image model itself is fine. Despite what some people want to

The woke nonsense isn't the source of the problem, it's a symptom of a deeper problem. The Gemeni image model itself is fine. Despite what some people want to believe, it wasn't trained to be anti-white. The problem was some people in high level management position decided they should insert DEI stuff into the prompt without asking or telling the users, and they forced the engineers to implement this without thinking through the obvious consequences, or even performing basic testing. A competently managed company could still make reasonable efforts to promote DEI (or not), but it would look nothing like the nonsensical images that Gemeni was putting out on launch. Jamming DEI in sideways where it doesn't belong and then having it blow up in their face is a symptom of deeper problems from the top. The DEI is just exacerbating the lack of clarity, which is a much bigger problem. Fixing Gemeni image generation is easy: just undo the stupid prompt insertions. Fixing the chain of decision-making authority that led to those prompt insertions is difficult. They need to restructure management.