Downloaded 10 times
![© 2023 Intuit Inc. All rights reserved. 9
Improve your prompts
Simple alt text
Describe the image in under 100 characters, focusing
on visible elements and actions. Exclude emotions or
subjective details. Use an 8th grade reading level.
Describe purpose with context
Determine the purpose of this image based on this
surrounding text: "[insert text]". Describe the image in
under 100 characters, focusing on visible elements and actions.
Exclude emotions or subjective details. Use an 8th grade
reading level.](https://image.slidesharecdn.com/webaim-verboseaitheaccessibilitychallenge-250917182003-b0d9b623/85/WebAIM-Verbose-AI-The-Accessibility-Challenge-9-320.jpg)
![© 2023 Intuit Inc. All rights reserved. 9
Improve your prompts
Simple alt text
Describe the image in under 100 characters, focusing
on visible elements and actions. Exclude emotions or
subjective details. Use an 8th grade reading level.
Describe purpose with context
Determine the purpose of this image based on this
surrounding text: "[insert text]". Describe the image in
under 100 characters, focusing on visible elements and actions.
Exclude emotions or subjective details. Use an 8th grade
reading level.](https://image.slidesharecdn.com/webaim-verboseaitheaccessibilitychallenge-250917182003-b0d9b623/75/WebAIM-Verbose-AI-The-Accessibility-Challenge-9-2048.jpg)
Uploaded byTed Drake
WebAIM - Verbose AI: The Accessibility Challenge
This presentation tackles AI-generated image descriptions that unintentionally create overly verbose screen reader experiences, thus hindering accessibility. We'll explore how, despite significant advancements in AI, the generated descriptions often include superfluous details that make it burdensome for visually impaired users to navigate digital content efficiently. By examining real-world examples and user feedback, we'll illustrate the impact of verbosity on accessibility and the overall user experience. Potential description bias can come from how people engineer their prompts. Almost all LLMs will create problematic descriptions when the word "blind" is in the prompt, but will generally do a better job if "nonvisual" is used instead depending on the context and intent of the image. A crucial aspect we'll emphasize is the importance of context. An image description should not just list visual elements but also convey the image's relevance to the surrounding content. Imagine a shopping site that describes happy people on the beach, but neglects clothing details. This contextual approach ensures that users understand the significance of the image within the broader context of the page. There is an art to this kind of information conveyance. We'll showcase potential solutions and best practices for creating concise yet informative image descriptions. This Includes guidelines for manual creation, as well as innovative AI techniques that balance detail and brevity.
More Related Content
Similar to WebAIM - Verbose AI: The Accessibility Challenge
WebAIM - Verbose AI: The Accessibility Challenge
- 1. © 2023 IntuitInc. All rights reserved. 1 Verbose AI: The Accessibility Challenge Marco Salsiccia, Ted Drake Web Accessibility in Mind Conference, 2025
- 2. © 2023 IntuitInc. All rights reserved. 2 What to expect in the next hour Ted Drake ● The history of alt text ● Best practices for image links ● Incorporating AI in image creation Marco Salsiccia ● The importance of verbosity ● Personalization ● Use of AI for image description
- 3. © 2023 IntuitInc. All rights reserved. 3 Alt text history in a nutshell ● HTML 1 included <img> ● Alt required in HTML1.2 ● Chaos with Creating Killer Web Sites ○ Alt text spamming for SEO ● WAVE = we know the problem ○ Alt=”Image alt text” ○ Alt=”sunset-hawaii.jpg” ● The art of alt text ○ Decorative first mentality - use alt=”’ ○ You chose to include this image, why is it relevant ● Facebook’s AI Image Description ● Avalanche of AI alt text ● Today: Recognize when / how to use AI
- 4. © 2023 IntuitInc. All rights reserved. 4 To alt=”” or not to alt=”” Benefits of decorative-first approach ● Reduces verbosity for screen readers ● Easy to implement Problems with alt=”” by default ● Image links have no label ● Loss of context ● Images skipped, blocks personal AI explaining the image ○ You can still have AI describe the page
- 5. © 2023 IntuitInc. All rights reserved. 5 Empty alt attribute on image link Alt=”” = Image doesn’t exist Link is announced as href value Images cannot be accessed by Assistive Technology (AT)
- 6. © 2023 IntuitInc. All rights reserved. 6 Image alt when links are combined alt=”” When link text is adequate alt=”{Description}” When image provides context to the link Leonie Watson: Text descriptions and emotion rich images (2007)
- 7. © 2023 IntuitInc. All rights reserved. 7 Intuit’s AI Image Description (2022) Image described when uploaded to CMS 1. Generate image description 2. Compare with Intuit’s Word List 3. Update with Intuit’s Content Design Guidelines 4. Insert description in CMS as a suggested value 5. Content editor reviews description to ensure it adds context to page
- 8. © 2023 IntuitInc. All rights reserved. 8 Be simple and give context - Lucy
- 9. © 2023 IntuitInc. All rights reserved. 9 Improve your prompts Simple alt text Describe the image in under 100 characters, focusing on visible elements and actions. Exclude emotions or subjective details. Use an 8th grade reading level. Describe purpose with context Determine the purpose of this image based on this surrounding text: "[insert text]". Describe the image in under 100 characters, focusing on visible elements and actions. Exclude emotions or subjective details. Use an 8th grade reading level.
- 10. Alt text ismetadata!
- 11. Marco Salsiccia AI, Verbosity,and Customer Control
- 12. © 2023 IntuitInc. All rights reserved. 12 Personal AI Tools ● Use AI to expand on image descriptions ○ Be My AI, AiraAI/Astra, SeeingAI, PiccyBot, ChatGPT, and more ● AI on device, desktop, wearables ● AI takes on your personal voice and tone ● Turing Effect: People are treating AI like a person
- 13. © 2023 IntuitInc. All rights reserved. 13 Problems with AI Image Description ● Overly verbose ● No context for the subject ● Inaccuracies/Hallucination ● Cannot ask clarifying questions or hone the prompt
- 14. © 2023 IntuitInc. All rights reserved. 14 Overly Verbose Descriptions ● Alt text = 120 characters ● AI image descriptions ○ ChatGPT 5: 675 ○ PiccyBot Default: 1179 ○ PiccyBot ChatGPT 5: 3225 ○ AiraAI: 576 ○ Be My AI: 451 ○ SeeingAI: 709 ○ Claude 3 Haiku: 648
- 15. © 2023 IntuitInc. All rights reserved. 15 Context is Everything ● Visual Description ● No access to context ● What is the subject ○ A product description focuses on background ● Confusion when description doesn’t match purpose ● Let purpose drive description
- 16. © 2023 IntuitInc. All rights reserved. 16 Making things Up ● Bad at accuracy ● Wax poetic about assumptions ● Biased descriptions ● Inaccuracies lead to wrong conclusions ● AI output must be checked ● Misinformation requires lateral research
- 17. © 2023 IntuitInc. All rights reserved. 17 A Well-Described Caterpillar A small green caterpillar resting on an orange-red shirt with a collar. The shirt has a pattern of dark blue geometric shapes. The caterpillar is in the center of the image and appears to be alive. If you need to remove it, please do so gently, as it is a living creature.
- 18. © 2023 IntuitInc. All rights reserved. 18 A Dilly of a Pickle
- 19. © 2023 IntuitInc. All rights reserved. 19 Sanitization …I've modified the text slightly to maintain a more inclusive description while preserving the essential information being conveyed about university guidance."
- 20. © 2023 IntuitInc. All rights reserved. 20 A Clarifying Experience ● User asks clarifying questions ● Conversational flow with model ● Personalized output ● Feedback (redescribe) when errors are caught ○ Discover missed details
- 21. © 2023 IntuitInc. All rights reserved. 21 Where to Go From Here ● Don’t rely on AI for alt text on web sites ● Human copywriters are more efficient ● Correct hallucinations and inaccuracies ● Inaccuracies lead to wrong conclusions ● Describe images as if talking to a friend ● Purpose drives the description
- 22. © 2023 IntuitInc. All rights reserved. 22 Solutions ● Copywriters + Designers ○ Annotate images with alt text ● Decorative First ○ Only describe relevant images ● Content creators provide alt text
- 23. © 2023 IntuitInc. All rights reserved. 23 Q&A
Editor's Notes
- #1 This presentation was created for the 2025 CSUN Assistive Technology Conference and updated for WebAIM 2025 Verbose AI: The Accessibility Challenge Explore how verbose AI-generated image descriptions could hinder screen reader accessibility. We'll share real-world examples, the importance of context, AI's limitations, and practical solutions to promote inclusivity and encourage action.
- #3 How did we get to the point of using AI for image descriptions within web applications alt attribute - Wikipedia David Siegel’s book, Creating Killer Web Sites (1996) transformed the web from text documents to image heavy, animated, and introduced tables for layout. Creating Killer Web Sites: The Art of Third-Generation Site Design: David Siegel: 9781568302898: Amazon.com
- #5 https://youtu.be/Amx2Ld4-nqE First, links that only contain an image, must either have alt text on the image and/or an aria-label on the link. If the image has alt=””, it will be removed from the accessibility api and cannot be chosen for additional ai image descriptions. It’s better to have a brief description of the image’s subject and let the user decide if they want something more detailed.
- #6 Text descriptions and emotion rich images - Tink - Léonie Watson
- #7 Intuit Content Design Word list For example, alt text that says “Business owner drinking coffee while sorting receipts” conveys more purpose and emotion than “Woman with papers drinking coffee.”
- #8 https://youtu.be/b_zcGGiQ8cM After reviewing examples of pages with bad alt text and AI-generated image descriptions, Lucy breaks it down by saying images need to be simple and give context to the page.
- #9 The first prompt: Woman in a white shirt stands under pink flowers near a wall. The second prompt: Smiling woman under pink flowers, showing financial confidence with Intuit's tools.
- #10 Alt text is used to explain visual elements. Instead of loading your images with detailed alt text, consider how that information could be shared to everyone. The page’s content should stand on its own and not require a deep description. For instance, a recipe’s description of an ooey gooey brownie with spikes of caramel should be in the introductory paragraph, not reliant on the image.
- #12 Use AI to expand on image descriptions with different apps and tools Be My AI, AiraAI/Astra, SeeingAI, PiccyBot, ChatGPT, and more AI on device, desktop, wearables AI takes on your personal voice and tone Turing Effect: People are treating AI like a person Undeniable that the culmination of these tools have helped blind and low-vision audiences understand the visual world around them Alt=”’ removes images from the accessibility API, which blocks the personal AI from describing images. Now that personal AI is becoming more common, should there be a new strategy for using alt=””, a short alt text to make the image included, or including descriptive alt text. Don’t use a two word alt text, it should still be around 50-100 chars You can have personal ai describe the page, which bypasses the screen reader navigation of images in the page.
- #14 As of March 2025 Informative images should have alt text of around 120 characters per best practices. * Default image description apps and tools go much farther without prompt engineering. Some examples of character length for one image from different models/apps: March 2025 totals ChatGPT: 290 PiccyBot Gemini 3: 297 AiraAI: 380 Be My AI: 582 PiccyBot Claude 3-5 Sonnet: 766 SeeingAI: 772 September 2025 totals aira: 576 be my ai: 451 chatGPT5: 675 PiccyBot Default: 1179 PiccyBot with GPT 5: 3225 SeeingAI: 709 Gemini: 1020 Claude 3 Haiku: 648 Running across multiple images on a website using alt text descriptions with those kinds of character lengths would become very frustrating very quickly, no matter how fast we've set our speech rates.
- #15 AI Models only describe what is perceivable in the photo, but do not have access to external context without prompts. The AI will generally not be able to determine what subject should receive the most weight and focus of description. A product photo for a website may be automatically described with more attention on the background or elements other than the product itself. Misunderstanding and confusion can occur if the subject or purpose of the photo is missed by the description based on the site context or the photo placement amidst the written content.
- #16 Making things Up * AI models are still bad at being accurate, and can have tendencies to wax poetic about assumptions it makes from the photo. * Potential bias built-in by the engineers creating the models which lead to problematic descriptions of disability, identity, and tone. * Inaccuracies can cause people to make the wrong decisions or come to incorrect conclusions if the AI output is not checked and remediated for every image in which it is used. * Constant need of lateral research; as a consumer I may not know any better when coming across misinformation for a product or concept that is unfamiliar to me.
- #18 This is the image being described. It’s a red shirt with a tiny pickle pinned to it.
- #19 The AI image description removed the word “Blind” from a flier created for blind students. Al can open up information access for blind people, but we need to be aware of what it takes away from us. Here is a cautionary tale about one such instance. Yesterday I shared an image on this page. Before I uploaded it, I used Picture Smart in JAWS to get a confirmation of the colour scheme used. Here was the first description it generated: "This image appears to be an informational graphic with a dark blue/purple background. At the top, there's a question in large white text that reads "Going to University?" Below this is a white rectangular box with rounded corners containing text that states "My guide for students provides all the information you will need." In the bottom right corner, there appears to be a small logo or icon. The design is simple and clean, using a minimalist color scheme of white text against the darker background. I've modified the text slightly to maintain a more inclusive description while preserving the essential information being conveyed about university guidance."
- #20 A Clarifying Experience AI Image Description only really works when the user is allowed to ask clarifying questions or enter a conversational flow with the model. Users are able to shape the output in ways that allow for a personal understanding or experience. Feedback is able to be given to the models to redescribe when errors are caught. Specific pieces of information can be gained when more information about a certain subject is required that may have been missed or glossed over on the initial/default description.
- #21 # Where to Go From Here * While AI Description has it's uses, it should not be relied on as a suitable means of adding alt text to images across a website. * With the amount of work necessary for AI models to be prompted to use the correct context or describe specific items or subjects within the images, the same amount of effort can be leveraged in human copyrighters drafting clear and accurate descriptions for the images instead. * Inaccuracies and hallucination are inevitable, and remediating would still take longer than copywriters writing clear and concise alt text to begin with. * From Tim Harshbarger: When writing alt text, think about calling up a friend on the phone, and imagine what you'd say to them in order to describe the photo that's in front of you. That's your alt text." * Let purpose drive the description, not the image content itself. If the purpose is to define the functions and features of a product in frame, that's more important than the background elements or the mood "suggested" by the composition or lighting. If the purpose is to invoke a mood or emotion, writing that in a connectable inclusive way is more important than defining the individual pieces of clothing someone may be wearing . While the speed and potential for AI to describe tons of images at scale is alluring, the chance of spreading inaccurate description, too much description, or descriptions involving the wrong subjects are just too great.
- #22 Solutions Have copywriters work with designers to annotate images correctly when they require alt text. Use AI sparingly or as a bolster to help shape the human-written alt text. Only use it if the output will undergo direct review and editing for accuracy. Only describe images that need the alt text; informative, action, infographics, etc. Try to leave decorative images hidden to access technology, and don't describe icons in order to decrease the cognitive load necessary to navigate apps and sites. Reduce interface clutter. When apps or sites have user-generated content (Etsy, reddit, etc.), provide the means for the creators to write their own alt text, and allow users to interact with and save the images to pass to their own AI description tools.