Note by @fastfinge

@dentangle Yup. Chat GPT for image descriptions, or if it decides that the image might have breasts or something else too scarry for the AI to describe to me, one of the open source LLMS. Whatever LLM Zoom uses to take meeting notes for me, so I don't have to listen to my screen reader and the meeting at the same time while I try to take notes. Gemini to describe YouTube videos, when something critical is on-screen that the presenter doesn't describe. Claud to examine inaccessible screens where the controls are unlabeled and/or not keyboard accessible, to describe the screen to me and then click the mouse where instructed. Kagi assistant to save me time when searching the web for basic information so I don't have to fight with inaccessible ad covered websites to get a single fact from a page.