The Gregarious Dragon
So, what do you all think of this image description prompt I use for AI tools? When I provide visual media (images, screenshots), act as a professional Audio Describer for the blind. Style: Use natural, concise prose. Focus on the main subject, action, and setting. Format: Start the description immediately. Do not use filler phrases like 'This image shows' or 'In this picture.' Screenshot Logic: Assess the image intent. Text-Heavy (Docs, Errors): Prioritize summarizing the text content and structure. Design-Heavy (Themes, UI, Layouts): Prioritize the visual aesthetics, describing color palettes, icon styles, and the overall 'look and feel'. #A11Y #Accessibility #Blind #AIPrompts #ImageDescriptions #Poll
I like it
75%
It could use some improvement (please explain)
0%
It's not for me
25%
Poll ended at .
Mar 23 at 10:15amWeb
Show threadToddZ ⓋMar 24

@somegregariousdude

https://ieji.de/@anantagd/115804706509841365

#BlindAltBot (@[email protected])

@[email protected] I am blind. Seeing people who think I'm not worth the effort fill my timeline with AltBot generated AI stuff that isn't even accurate in lots of cases. Human alt text is always better, because it doesn´t focus on ocular seeing. Seeing people think, and AltBot was designed around that notion, that blind people must compensate for missing "eye-seeing", but that's not the case. I am interested in the meaning of an image to you, its maker or publisher. Again, human alt text is better, also because it strengthens reciprocity between seeing and blind people. AltBot doesn't but it makes seeing people believe they have done their bit for accessibility. In actuality, the reverse is often true. !!!!!!!!!

ieji.de
Show threadThe Gregarious DragonMar 24
@toddz I'm blind myself and would have no other way to get image descriptions since I can't easily ask a fellow human to describe something to me. Besides, the issue with having a fellow human provide a description is that they might not describe the image in the detail I need to get a clear idea of what the image represents. Using AI sidesteps the issue for me, as I can specify how I'd like the image described, and I'd usually get a consistent result.
Show threadToddZ ⓋMar 25
@somegregariousdude Ah, ok. I admit I did not imagine that use case.
Show threadBrandon TysonMar 24
@somegregariousdude Is this one prompt that you're using across the board? I'm having trouble following this and I just want to make sure I'm understanding right.
Show threadThe Gregarious DragonMar 24
@BTyson I use this prompt with the Gemini AI plugin for NVDA as well as with Claude and Alexa Plus. It's based on a lot of experimentation and generally gives me the information I want without the fluff I don't.