Opportunities for AI in Accessibility  ## Metadata - Author: [[Tom Shannon]] - Full Title: Opportunities for AI in Accessibility - Category: #articles - Document Note: --- tags: [#AI, #Accessibility, #Inclusion, #EthicalAI, #VoicePreservation] Author: [Author's Name] Source: [Blog/Website] Reference: [Link] date created: 2024-09-11 10:30 --- # Title: AI and Accessibility: Opportunities and Risks ## Table of Contents: 1. [Introduction](#Introduction) 2. [AI for Accessibility](#AI-for-Accessibility) 3. [Computer Vision and Alt Text](#Computer-Vision-and-Alt-Text) 4. [Opportunities for People with Disabilities](#Opportunities-for-People-with-Disabilities) 5. [The Role of Inclusive Algorithms](#The-Role-of-Inclusive-Algorithms) 6. [Challenges with AI Bias and Diversity](#Challenges-with-AI-Bias-and-Diversity) 7. [Conclusion](#Conclusion) ### References: - Noble, Safiya Umoja, "Algorithms of Oppression" - Microsoft AI for Accessibility grant program ## 📝 Summary: This article discusses the potential benefits and risks of using AI to improve accessibility for people with disabilities. While acknowledging skepticism about AI's current limitations, the author explores various positive applications, such as alt text generation, voice preservation, and inclusive matching algorithms. The piece emphasizes the need for diverse teams in AI development to avoid harmful biases and outlines how AI could evolve to better serve marginalized communities. ## 📖 Main Content: ### AI for Accessibility: - AI holds transformative potential to assist people with disabilities, especially in areas like voice recognition and computer vision. - The AI for Accessibility grant program is working to identify projects that can use AI to provide tangible improvements in the lives of people with disabilities. ### Computer Vision and Alt Text: - Current alt text generation models, while improving, still lack contextual understanding, often leading to poor descriptions. - The potential exists for AI to assist in labeling decorative vs. essential images and provide richer, context-sensitive alt text. ### Opportunities for People with Disabilities: - **Voice Preservation**: AI models can replicate voices, which is crucial for individuals with ALS or motor-neuron diseases. - **Voice Recognition**: Ongoing research aims to make voice-based systems more inclusive for people with atypical speech patterns. - **Text Simplification**: LLMs can adjust content to cater to people with cognitive disabilities, offering simplified summaries or versions. ### The Role of Inclusive Algorithms: - **Mentra Case Study**: Mentra, an employment network for neurodivergent individuals, uses inclusive algorithms to match job seekers with employers. This algorithm focuses on both the candidate’s strengths and required accommodations. - Algorithms tuned to reduce bias can play a significant role in building more inclusive communities. ### Challenges with AI Bias and Diversity: - Diverse teams are essential in reducing harmful biases in AI. - Without diversity in the data and creators, AI models risk perpetuating ableist or exclusionary practices. - Diverse data sets are key to developing models that promote accessibility, like coding copilots that encourage accessible design. ## 🗒️ Notes: - AI can be both inclusive and exclusionary depending on how it is built. - AI systems still struggle with the context of images, especially for generating alt text. - There's potential to improve accessibility tools through AI-driven solutions like voice replication and text simplification. - Diverse data and teams are essential to reducing bias in AI algorithms. ## 📖 References: - Noble, Safiya Umoja, "Algorithms of Oppression" - Microsoft AI for Accessibility program ## 🗣️ Paraphrase: 1. *Original*: "Computer-vision models continue to improve in the quality and richness of detail in their descriptions, but their results aren’t great... They’re not trained to distinguish between images that are contextually relevant and those that are purely decorative." - **Paraphrase**: While AI image recognition is improving, it still struggles to provide meaningful descriptions, especially when distinguishing between important and decorative images. 2. *Original*: "When more people with disabilities are involved in the creation of algorithms, that can reduce the chances that these algorithms will inflict harm on their communities." - **Paraphrase**: Including more individuals with disabilities in the creation of AI algorithms helps minimize the risk of these systems causing harm to those communities. ## 📰 Newsletter: This week, we're exploring how AI could revolutionize accessibility. From voice preservation for people with ALS to smarter algorithms for neurodivergent job seekers, there are exciting developments ahead. But with great power comes great responsibility—without diverse voices in the room, AI could just as easily deepen divides. Learn more about the potential for AI in accessibility and how inclusive design is key to making tech work for everyone. Check out the full article [here]. ## ✅ Action Items: This article directly aligns with my professional focus on fostering inclusive open-source communities and advocating for accessible technologies. AI can enhance contributor engagement, particularly for underrepresented groups, by offering tools like voice recognition or cognitive-friendly text simplification. By promoting diversity in the development of AI models, we can ensure that accessibility is built into the foundation of tech solutions, which ties into my commitment to an inclusive WordPress ecosystem. ## 🧪 Hypotheses: 1. Incorporating more diverse voices in AI development teams will lead to a measurable decrease in harmful biases in AI tools. 2. Voice preservation technology will become a standard accessibility feature for people with degenerative diseases within five years. 3. AI-driven tools like simplified text or image-to-text conversion will increase accessibility in educational content for individuals with cognitive disabilities. ### 📊 Methodology: - Conduct surveys of users with disabilities on their experience using AI-based accessibility tools (e.g., voice recognition, alt text). - Analyze the performance of AI models trained on diverse datasets vs. less diverse datasets. - Track adoption rates of AI-based accessibility tools in various sectors (education, employment, healthcare). ### ❓ Questions: - How can we ensure that diverse datasets are used in AI development? - What are the ethical implications of voice preservation technology, particularly concerning deepfakes? - How can AI-driven solutions balance efficiency with inclusivity? ### 🔮 Future Research: - Explore the long-term effects of AI on accessibility in education. - Investigate how AI can help identify and mitigate biases in real-time content generation. - Study the integration of AI tools in public and private sector accessibility programs. ### 💡 Implications: - AI could revolutionize accessibility, but its development must be guided by inclusive principles to ensure it benefits all users, particularly those with disabilities. - Focusing on diverse data sets and teams will be critical to mitigating biases that disproportionately affect marginalized groups. - Ethical concerns, especially around deepfakes and bias, need to be addressed to build trust in AI-powered accessibility tools. ### ✅ Logical Validity: The article presents well-supported arguments for the positive impact AI can have on accessibility, while also highlighting the risks. The tone is balanced, acknowledging both the potential and the challenges of AI. The points about the importance of diverse teams are particularly strong and supported by examples like Mentra’s algorithm for neurodivergent job seekers. **Sound Reasoning**: - The author’s argument that diverse teams reduce bias is well supported with examples and aligns with current discussions on ethical AI. **Suggestions for Improvement**: - More data on the success rates of AI-driven accessibility tools would strengthen the argument. **Logical Fallacies**: None detected. ## 🎯 Conclusion: AI holds tremendous promise for improving accessibility, but it must be developed with inclusivity and diversity at its core. From alt text to voice recognition, AI can make technology more accessible for people with disabilities, but we must be cautious of the biases these systems may inherit without diverse teams and data sets. - Document Tags: [[ai]] [[joe dolson]] [[accessibility]] - Summary: AI has the potential to make meaningful differences for people with disabilities in various ways. For example, AI can be used to improve the generation of alternative text for images, helping to provide more accurate and contextual descriptions. AI algorithms can also match job seekers with potential employers based on their strengths and necessary accommodations. When diverse teams are involved in creating algorithms, it reduces the chances of harm being inflicted on marginalized communities. AI can also be used for voice preservation, voice recognition, text transformation, and more to benefit individuals with disabilities. It is important to include diverse perspectives and lived experiences in the development of AI models and data sets to ensure more equitable outcomes. - URL: https://alistapart.com/article/opportunities-for-ai-in-accessibility/ In reading [Joe Dolson’s recent piece on the intersection of AI and accessibility](https://www.joedolson.com/2023/06/accessibility-and-artificial-intelligence/), I absolutely appreciated the skepticism that he has for AI in general as well as for the ways that many have been using it. In fact, I’m very skeptical of AI myself, despite my role at Microsoft as an accessibility innovation strategist who helps run the AI for Accessibility grant program. As with any tool, AI can be used in very constructive, inclusive, and accessible ways; and it can also be used in destructive, exclusive, and harmful ones. And there are a ton of uses somewhere in the mediocre middle as well. Article Continues Below I’d like you to consider this a “yes… and” piece to complement Joe’s post. I’m not trying to refute any of what he’s saying but rather provide some visibility to projects and opportunities where AI can make meaningful differences for people with disabilities. To be clear, I’m not saying that there aren’t real risks or pressing issues with AI that need to be addressed—there are, and we’ve needed to address them, like, yesterday—but I want to take a little time to talk about what’s possible in hopes that we’ll get there one day. Joe’s piece spends a lot of time talking about computer-vision models generating alternative text. He highlights a ton of valid issues with the current state of things. And while computer-vision models continue to improve in the quality and richness of detail in their descriptions, their results aren’t great. As he rightly points out, the current state of image analysis is pretty poor—especially for certain image types—in large part because current AI systems examine images in isolation rather than within the contexts that they’re in (which is a consequence of having separate “foundation” models for text analysis and image analysis). Today’s models aren’t trained to distinguish between images that are contextually relevant (that should probably have descriptions) and those that are purely decorative (which might not need a description) either. Still, I still think there’s potential in this space. As Joe mentions, human-in-the-loop authoring of alt text should absolutely be a thing. And if AI can pop in to offer a starting point for alt text—even if that starting point might be a prompt saying *What is this BS? That’s not right at all… Let me try to offer a starting point*—I think that’s a win. Taking things a step further, if we can specifically train a model to analyze image usage in context, it could help us more quickly identify which images are likely to be decorative and which ones likely require a description. That will help reinforce which contexts call for image descriptions *and* it’ll improve authors’ efficiency toward making their pages more accessible. While complex images—like graphs and charts—are challenging to describe in any sort of succinct way (even for humans), [the image example shared in the GPT4 announcement](https://openai.com/research/gpt-4#:~:text=Visual%20inputs:%20VGA%20charger) points to an interesting opportunity as well. Let’s suppose that you came across a chart whose description was simply the title of the chart and the kind of visualization it was, such as: *Pie chart comparing smartphone usage to feature phone usage among US households making under $30,000 a year.* (That would be a pretty awful alt text for a chart since that would tend to leave many questions about the data unanswered, but then again, let’s suppose that that was the description that was in place.) If your browser knew that that image was a pie chart (because an onboard model concluded this), imagine a world where users could ask questions like these about the graphic: * *Do more people use smartphones or feature phones?* * *How many more?* * *Is there a group of people that don’t fall into either of these buckets?* * *How many is that?* Setting aside the realities of [large language model (LLM) hallucinations](https://machinelearningmastery.com/a-gentle-introduction-to-hallucinations-in-large-language-models/)—where a model just makes up plausible-sounding “facts”—for a moment, the opportunity to learn more about images and data in this way could be revolutionary for blind and low-vision folks as well as for people with various forms of color blindness, cognitive disabilities, and so on. It could also be useful in educational contexts to help people who *can* see these charts, as is, to understand the data in the charts. Taking things a step further: What if you could ask your browser to simplify a complex chart? What if you could ask it to isolate a single line on a line graph? What if you could ask your browser to transpose the colors of the different lines to work better for form of color blindness you have? What if you could ask it to swap colors for patterns? Given these tools’ chat-based interfaces and our existing ability to manipulate images in today’s AI tools, that seems like a possibility. Now imagine a purpose-built model that could extract the information from that chart and convert it to another format. For example, perhaps it could turn that pie chart (or better yet, a series of pie charts) into more accessible (and useful) formats, like spreadsheets. That would be amazing! #### **Matching algorithms**<#section3> Safiya Umoja Noble absolutely hit the nail on the head when she titled her book [*Algorithms of Oppression*](http://algorithmsofoppression.com/). While her book was focused on the ways that search engines reinforce racism, I think that it’s equally true that all computer models have the potential to amplify conflict, bias, and intolerance. Whether it’s Twitter always showing you the latest tweet from a bored billionaire, YouTube sending us into a Q-hole, or Instagram warping our ideas of what natural bodies look like, we know that poorly authored and maintained algorithms are incredibly harmful. A lot of this stems from a lack of diversity among the people who shape and build them. When these platforms are built with inclusively baked in, however, there’s real potential for algorithm development to help people with disabilities. Take [Mentra](https://www.mentra.com/), for example. They are an employment network for neurodivergent people. They use an algorithm to match job seekers with potential employers based on over 75 data points. On the job-seeker side of things, it considers each candidate’s strengths, their necessary and preferred workplace accommodations, environmental sensitivities, and so on. On the employer side, it considers each work environment, communication factors related to each job, and the like. As a company run by neurodivergent folks, Mentra made the decision to flip the script when it came to typical employment sites. They use their algorithm to propose available candidates to companies, who can then connect with job seekers that they are interested in; reducing the emotional and physical labor on the job-seeker side of things. When more people with disabilities are involved in the creation of algorithms, that can reduce the chances that these algorithms will inflict harm on their communities. That’s why diverse teams are so important. Imagine that a social media company’s recommendation engine was tuned to analyze who you’re following and if it was tuned to prioritize follow recommendations for people who talked about similar things but who were different in some key ways from your existing sphere of influence. For example, if you were to follow a bunch of nondisabled white male academics who talk about AI, it could suggest that you follow academics who are disabled or aren’t white or aren’t male who also talk about AI. If you took its recommendations, perhaps you’d get a more holistic and nuanced understanding of what’s happening in the AI field. These same systems should also use their understanding of biases about particular communities—including, for instance, the disability community—to make sure that they aren’t recommending any of their users follow accounts that perpetuate biases against (or, worse, spewing hate toward) those groups. #### **Other ways that AI can helps people with disabilities**<#section4> If I weren’t trying to put this together between other tasks, I’m sure that I could go on and on, providing all kinds of examples of how AI could be used to help people with disabilities, but I’m going to make this last section into a bit of a lightning round. In no particular order: * Voice preservation. You may have seen [the VALL-E paper](https://arxiv.org/abs/2301.02111) or [Apple’s Global Accessibility Awareness Day announcement](https://www.apple.com/newsroom/2023/05/apple-previews-live-speech-personal-voice-and-more-new-accessibility-features/) or you may be familiar with the voice-preservation offerings from [Microsoft](https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/custom-neural-voice), [Acapela](https://www.acapela-group.com/voices/voice-banking/), or others. It’s possible to train an AI model to replicate your voice, which can be a tremendous boon for people who have ALS (Lou Gehrig’s disease) or motor-neuron disease or other medical conditions that can lead to an inability to talk. This is, of course, the same tech that can also be used to create audio deepfakes, so it’s something that we need to approach [*responsibly*](https://www.microsoft.com/en-us/ai/responsible-ai), but the tech has truly transformative potential. * Voice recognition. Researchers like those in the [Speech Accessibility Project](https://speechaccessibilityproject.beckman.illinois.edu/) are paying people with disabilities for their help in collecting recordings of people with atypical speech. As I type, they are actively recruiting people with Parkinson’s and related conditions, and they have plans to expand this to other conditions as the project progresses. This research will result in more inclusive data sets that will let more people with disabilities use voice assistants, dictation software, and voice-response services as well as control their computers and other devices more easily, using only their voice. * Text transformation. The current generation of LLMs is quite capable of adjusting existing text content without injecting hallucinations. This is hugely empowering for people with cognitive disabilities who may benefit from text summaries or simplified versions of text or even text that’s prepped for [Bionic Reading](https://bionic-reading.com/). #### **The importance of diverse teams and data**<#section5> We need to recognize that our differences matter. Our lived experiences are influenced by the intersections of the identities that we exist in. These lived experiences—with all their complexities (and joys and pain)—are valuable inputs to the software, services, and societies that we shape. Our differences need to be represented in the data that we use to train new models, and the folks who contribute that valuable information need to be compensated for sharing it with us. Inclusive data sets yield more robust models that foster more equitable outcomes. Want a model that doesn’t demean or patronize or objectify people with disabilities? Make sure that you have content about disabilities that’s authored by people with a range of disabilities, and make sure that that’s well represented in the training data. Want a model that doesn’t use ableist language? You may be able to use [existing data sets](https://www.selfdefined.app/) to build a filter that can intercept and remediate ableist language before it reaches readers. That being said, when it comes to sensitivity reading, AI models won’t be replacing human copy editors anytime soon.  Want a coding copilot that gives you accessible recommendations from the jump? Train it on code that you know to be accessible. --- I have no doubt that AI can and will harm people… today, tomorrow, and well into the future. But I also believe that we can acknowledge that and, with an eye towards accessibility (and, more broadly, inclusion), make thoughtful, considerate, and intentional changes in our approaches to AI that will reduce harm over time as well. Today, tomorrow, and well into the future. --- Many thanks to Kartik Sawhney for helping me with the development of this piece, Ashley Bischoff for her invaluable editorial assistance, and, of course, Joe Dolson for the prompt., Opportunities for AI in Accessibility, Tom Shannon, 01hpfn35cgata9hwr7mbk6f2c5.