{"id":1401,"date":"2024-10-10T11:29:25","date_gmt":"2024-10-10T11:29:25","guid":{"rendered":"https:\/\/html5accessibility.com\/stuff\/?p=1401"},"modified":"2024-10-10T14:26:54","modified_gmt":"2024-10-10T14:26:54","slug":"ai-for-accessibility-conformance-testing-why","status":"publish","type":"post","link":"https:\/\/html5accessibility.com\/stuff\/2024\/10\/10\/ai-for-accessibility-conformance-testing-why\/","title":{"rendered":"AI for accessibility conformance testing? Why?"},"content":{"rendered":"

\"A<\/p>\n

I have noticed a number of people are advocating the use of AI\/LLMs (Large Language Models) for accessibility testing and\/or code remediation.<\/p>\n

My onging question is why?<\/h2>\n

My personal experience with ChatGPT, for example is that it suck<\/em>s at these tasks. Or it may well be that I suck<\/strong> at using AI to elicit useful\/trustworthy results or safe accessible code. What tricks, prompt engineering skills am I missing?<\/p>\n

an aside<\/h2>\n

One accessibility corp. that provided an AI powered “fix my code” site. Have taken it down, probably because it sucked<\/em>. But are still peddling AI code remediation as a solution<\/a><\/p>\n

An other aside<\/h3>\n

One of the uses of AI that I do find worthwhile is the generation of initial drafts<\/strong> of alt text for my article images, for example I asked chatGPT to describe the image in this article. The output:<\/p>\n

A surreal image of people wearing knitted full-body masks, \r\nwith text overlays reading \"For accessibility testing\" and \r\n\"For accessibility code remediation.\" \r\nTwo figures on the right are marked with large red Xs, \r\nsuggesting disapproval. The word \"AI\" appears in the upper \r\nleft, indicating the role of artificial intelligence in \r\naccessibility.<\/pre>\n
Which is not a bad start, but it’s easy to not notice the inaccuracies, which I did not in the first instance “Two figures on the right are marked with large red Xs, suggesting disapproval.” This part of the alt is incorrect as the X’s pertain to the text “For accessibility testing X” and “For accessibility code remediation. X” they are not placed where they are to obscure anything.<\/p>\n
After realising the mistake I updated the text to make it more accurate. note I also clarified the subject of the slide “indicating the role of artificial intelligence in accessibility testing\/remediation”<\/p>\n
A surreal image of people wearing knitted full-body masks, with \r\ntext overlays reading \"For accessibility testing\" and \r\n\"For accessibility code remediation.\" each is marked with a large\r\nred X, suggesting disapproval. The word \"AI\" appears in the upper \r\nleft, indicating the role of artificial intelligence in \r\naccessibility testing\/remediation.<\/pre>\n
Further reading<\/h2>\n