AI and Accessibility: We Got a Lot to Talk About

Rate this content
Bookmark
Slides

In the rapidly evolving landscape of technology, Artificial Intelligence (AI) stands out as a game-changer our industry. But what are the implications for accessibility and building compliant and inclusive experiences?

In this talk, we will discuss AI and accessibility in all its nuances. From significant advancements that improve experiences for all users to major concerns around the ethics and organizational investments in accessibility, we will cover it all. Either way, AI is here and we got a lot to talk about to ensure accessibility remains at the forefront of our work.

This talk has been presented at React Summit US 2024, check out the latest edition of this React Conference.

FAQ

AI is defined as a simulation of human intelligence by a machine. It ranges from simple chatbots to more complex systems like those in sci-fi movies. The presentation focuses on narrow AI, which performs specific tasks.

The presentation discusses three main types of AI: Narrow AI, General AI, and Super Intelligent AI, with a focus on Narrow AI, which is designed for specific tasks.

AI is used to improve accessibility by enhancing writing tools like Chat GPT for people with disabilities, improving text-to-speech applications, voice recognition, captioning, and translation tools, and aiding visually impaired users with image recognition technologies.

Hiring accessibility experts ensures that websites and applications are built with compliance and accessibility in mind, providing the best experience for all users, including those with disabilities.

Recommended tools for improving web accessibility include the Axe Core Browser Extension, Accessibility Insights for the Web, Lighthouse, Selenium from Deque, and Plywright accessibility extensions.

Engineers can improve web accessibility by cleaning up HTML, using accessibility browser tools, and integrating automated testing into the development pipeline to catch accessibility issues before production.

AI helps in creating accessible educational content by improving interaction tools and generating accessible code suggestions. However, it is not a complete solution and requires collaboration with accessibility experts.

AI does not replace human experts and can perpetuate societal biases. It may also produce inaccurate results and users might trust these results too much. AI is a tool that should be used strategically and with caution.

'Be My Eyes' is an AI application that helps visually impaired users by describing their surroundings using AI models, thus helping them navigate the world more independently.

Narrow AI is significant because it represents the majority of AI applications currently in use, performing specific tasks like self-driving cars and face recognition, which are crucial for accessibility advancements.

Cat Johnson
Cat Johnson
29 min
19 Nov, 2024

Comments

Sign in or register to post your comment.
Video Summary and Transcription
Let's start diving into the presentation on AI and accessibility. AI is a simulation of human intelligence by machines, and this presentation will focus on narrow AI. Generative AI has achieved great things in accessibility, and AI advancements have improved various applications for people with disabilities. OpenAI's newest model allows blind people to access AI through Be My Eyes. Challenges with AI include bias and accuracy. AI can be a useful tool in improving accessibility, but it has limitations and risks. Advocating for AI ethics and accessibility, collecting data, and addressing specific accessibility concerns are important. Khan Academy is working on improving accessibility in education.

1. Introduction to AI and Accessibility

Short description:

Let's start diving into the presentation on AI and accessibility. I am an accessibility expert with experience at Microsoft and Khan Academy. Today, I will discuss the impact of AI on accessibility, its limitations, pros, and cons. AI is a simulation of human intelligence by machines. There are three types of AI: narrow AI, general AI, and super intelligent AI. This presentation will focus on narrow AI, which performs specific tasks and is widely used in various applications.

Let's start diving into the presentation because we only have 20 minutes to talk about everything. As Daphne said, I'm Kat Johnson and today I'm going to be talking to you about AI and accessibility and there's a lot to talk about and we only have so much time.

So one thing I want to give a disclaimer of in this presentation is that I am not an AI expert. I am an accessibility expert. Most of my career has been working at Microsoft and now I work at Khan Academy where I specialize in making accessible web components and widgets that work with screen readers and other assistive tools.

So you're probably wondering why I'm talking about AI. Well, it's mostly because everyone asks me all the time. All the time people are asking me about AI and whether it's going to be this revolutionary tool that's going to fix all of our problems with accessibility. And really this presentation is my way of trying to answer that question, get it recorded, so I don't have to keep answering it all the time.

So today I really want to talk about all the ins and outs of AI as it relates to accessibility. I want to talk about what AI can do for accessibility. What are the limitations of AI? And what are the pros, cons, and everything else in between? So with that, let's start diving in and looking at the nuances of AI and accessibility.

But first, let's talk about what is AI? Because grounding ourselves in what is AI will help us most in answering this question. So what is AI? Well, this quote I found online, and I really this is my favorite quote, hence why it's in this presentation. AI is simply a simulation of human intelligence by a machine. This we see all over the place within our society from really interesting little AI chat bots when you're trying to update your credit card details to the fantasy of Terminator and Skynet taking over the world.

So that's a broad range of AI. So let's break it down to some actual terms that we can use to define what AI is. And really there's three types of AI that I'm going to touch upon. There's actually a lot more terms around AI. Obviously an AI expert could probably go into more detail. But for this presentation, I want to point out three main types. Narrow AI, general AI, and super intelligent AI.

Now this presentation could easily dive into each one of these subtype areas. But that would eat up all of our time. For this presentation, we're really just going to focus on narrow AI and what it is exactly. So narrow AI is really a type of AI that performs specific tasks and doesn't learn anything beyond that. One thing to note is that most of the AI that we're seeing out in the industry right now that we're all really excited about is just narrow AI. And you see narrow AI in so many different applications from self-driving cars to face recognition software to even those cool chess-playing robots that beat grandmasters at chess. But all these are narrow AI in application.

2. Generative AI and Achievements in Accessibility

Short description:

Generative AI is an extension of narrow AI that generates content based on large amounts of data. Examples include live-chat GPT, Copilot, Midjourney, and Synthesia. Accessibility focuses on enabling access to services and tools for people with disabilities. AI has greatly facilitated writing through tools like Chat GPT.

Now some of you are probably curious about generative AI, LLMs and those large language models, how do they work and what type of AI are they? Well generative AI is really just a subset, an extension of narrow AI with the purpose of generating content based on large amounts of data that it's been trained on. So it's still narrow AI. It's meant to do a very targeted set of things, but just reproduce it by generating content.

And there's a lot of content out in the market that's doing that today, live-chat GPT, Copilot, applying it in different ways, Midjourney for generating images, and Synthesia. I'm glad I said that correctly. But Synthesia generates videos. I haven't tested it out, but it was really interesting, so I included it in this presentation. So that's AI, and this presentation is going to be focusing on narrow AI.

Let's pivot slightly and talk about accessibility. And what is accessibility just to define it for us? Well accessibility simply means the ability to access. It typically focuses on enabling access to societal and community services and tools for people with disabilities. You can think about accessibility in terms of the closed captioning you see on your Netflix shows. Or the powered accessibility doors when you get into this conference center. And a lot of developers, especially us web developers, when we're working with and trying to make our code accessible, we're usually focusing on ensuring that our code can be picked up by different accessible tools such as click... Come on. There. Oops. It's glitching out. One second. Boop! All right. Braille readers. Yes. One of the tools I have to work on at work is ensuring that my website will work with Braille readers and screen readers such as NVIDIA, JAWS, VoiceOver, and so many more. But these are some good grounding things to think about when we're talking about accessibility.

Now let's move on to the exciting part of the presentation. Talking about all the achievements in accessibility with the help of AI. One big thing I want to make a note of is that AI has made writing so much easier for so many people. Chat GPT, a tool to help us write essays, write letters and emails. I don't know about you, but I detest writing emails. And so Chat GPT has been very helpful for me.

QnA

Check out more articles and videos

We constantly think of articles and videos that might spark Git people interest / skill us up or help building a stellar career

Don't Solve Problems, Eliminate Them
React Advanced 2021React Advanced 2021
39 min
Don't Solve Problems, Eliminate Them
Top Content
Kent C. Dodds discusses the concept of problem elimination rather than just problem-solving. He introduces the idea of a problem tree and the importance of avoiding creating solutions prematurely. Kent uses examples like Tesla's electric engine and Remix framework to illustrate the benefits of problem elimination. He emphasizes the value of trade-offs and taking the easier path, as well as the need to constantly re-evaluate and change approaches to eliminate problems.
Using useEffect Effectively
React Advanced 2022React Advanced 2022
30 min
Using useEffect Effectively
Top Content
Today's Talk explores the use of the useEffect hook in React development, covering topics such as fetching data, handling race conditions and cleanup, and optimizing performance. It also discusses the correct use of useEffect in React 18, the distinction between Activity Effects and Action Effects, and the potential misuse of useEffect. The Talk highlights the benefits of using useQuery or SWR for data fetching, the problems with using useEffect for initializing global singletons, and the use of state machines for handling effects. The speaker also recommends exploring the beta React docs and using tools like the stately.ai editor for visualizing state machines.
Design Systems: Walking the Line Between Flexibility and Consistency
React Advanced 2021React Advanced 2021
47 min
Design Systems: Walking the Line Between Flexibility and Consistency
Top Content
The Talk discusses the balance between flexibility and consistency in design systems. It explores the API design of the ActionList component and the customization options it offers. The use of component-based APIs and composability is emphasized for flexibility and customization. The Talk also touches on the ActionMenu component and the concept of building for people. The Q&A session covers topics such as component inclusion in design systems, API complexity, and the decision between creating a custom design system or using a component library.
React Concurrency, Explained
React Summit 2023React Summit 2023
23 min
React Concurrency, Explained
Top Content
Watch video: React Concurrency, Explained
React 18's concurrent rendering, specifically the useTransition hook, optimizes app performance by allowing non-urgent updates to be processed without freezing the UI. However, there are drawbacks such as longer processing time for non-urgent updates and increased CPU usage. The useTransition hook works similarly to throttling or bouncing, making it useful for addressing performance issues caused by multiple small components. Libraries like React Query may require the use of alternative APIs to handle urgent and non-urgent updates effectively.
Managing React State: 10 Years of Lessons Learned
React Day Berlin 2023React Day Berlin 2023
16 min
Managing React State: 10 Years of Lessons Learned
Top Content
Watch video: Managing React State: 10 Years of Lessons Learned
This Talk focuses on effective React state management and lessons learned over the past 10 years. Key points include separating related state, utilizing UseReducer for protecting state and updating multiple pieces of state simultaneously, avoiding unnecessary state syncing with useEffect, using abstractions like React Query or SWR for fetching data, simplifying state management with custom hooks, and leveraging refs and third-party libraries for managing state. Additional resources and services are also provided for further learning and support.
TypeScript and React: Secrets of a Happy Marriage
React Advanced 2022React Advanced 2022
21 min
TypeScript and React: Secrets of a Happy Marriage
Top Content
React and TypeScript have a strong relationship, with TypeScript offering benefits like better type checking and contract enforcement. Failing early and failing hard is important in software development to catch errors and debug effectively. TypeScript provides early detection of errors and ensures data accuracy in components and hooks. It offers superior type safety but can become complex as the codebase grows. Using union types in props can resolve errors and address dependencies. Dynamic communication and type contracts can be achieved through generics. Understanding React's built-in types and hooks like useState and useRef is crucial for leveraging their functionality.

Workshops on related topic

React Performance Debugging Masterclass
React Summit 2023React Summit 2023
170 min
React Performance Debugging Masterclass
Top Content
Featured WorkshopFree
Ivan Akulov
Ivan Akulov
Ivan’s first attempts at performance debugging were chaotic. He would see a slow interaction, try a random optimization, see that it didn't help, and keep trying other optimizations until he found the right one (or gave up).
Back then, Ivan didn’t know how to use performance devtools well. He would do a recording in Chrome DevTools or React Profiler, poke around it, try clicking random things, and then close it in frustration a few minutes later. Now, Ivan knows exactly where and what to look for. And in this workshop, Ivan will teach you that too.
Here’s how this is going to work. We’ll take a slow app → debug it (using tools like Chrome DevTools, React Profiler, and why-did-you-render) → pinpoint the bottleneck → and then repeat, several times more. We won’t talk about the solutions (in 90% of the cases, it’s just the ol’ regular useMemo() or memo()). But we’ll talk about everything that comes before – and learn how to analyze any React performance problem, step by step.
(Note: This workshop is best suited for engineers who are already familiar with how useMemo() and memo() work – but want to get better at using the performance tools around React. Also, we’ll be covering interaction performance, not load speed, so you won’t hear a word about Lighthouse 🤐)
React Hooks Tips Only the Pros Know
React Summit Remote Edition 2021React Summit Remote Edition 2021
177 min
React Hooks Tips Only the Pros Know
Top Content
Featured Workshop
Maurice de Beijer
Maurice de Beijer
The addition of the hooks API to React was quite a major change. Before hooks most components had to be class based. Now, with hooks, these are often much simpler functional components. Hooks can be really simple to use. Almost deceptively simple. Because there are still plenty of ways you can mess up with hooks. And it often turns out there are many ways where you can improve your components a better understanding of how each React hook can be used.You will learn all about the pros and cons of the various hooks. You will learn when to use useState() versus useReducer(). We will look at using useContext() efficiently. You will see when to use useLayoutEffect() and when useEffect() is better.
React, TypeScript, and TDD
React Advanced 2021React Advanced 2021
174 min
React, TypeScript, and TDD
Top Content
Featured WorkshopFree
Paul Everitt
Paul Everitt
ReactJS is wildly popular and thus wildly supported. TypeScript is increasingly popular, and thus increasingly supported.

The two together? Not as much. Given that they both change quickly, it's hard to find accurate learning materials.

React+TypeScript, with JetBrains IDEs? That three-part combination is the topic of this series. We'll show a little about a lot. Meaning, the key steps to getting productive, in the IDE, for React projects using TypeScript. Along the way we'll show test-driven development and emphasize tips-and-tricks in the IDE.
Designing Effective Tests With React Testing Library
React Summit 2023React Summit 2023
151 min
Designing Effective Tests With React Testing Library
Top Content
Featured Workshop
Josh Justice
Josh Justice
React Testing Library is a great framework for React component tests because there are a lot of questions it answers for you, so you don’t need to worry about those questions. But that doesn’t mean testing is easy. There are still a lot of questions you have to figure out for yourself: How many component tests should you write vs end-to-end tests or lower-level unit tests? How can you test a certain line of code that is tricky to test? And what in the world are you supposed to do about that persistent act() warning?
In this three-hour workshop we’ll introduce React Testing Library along with a mental model for how to think about designing your component tests. This mental model will help you see how to test each bit of logic, whether or not to mock dependencies, and will help improve the design of your components. You’ll walk away with the tools, techniques, and principles you need to implement low-cost, high-value component tests.
Table of contents- The different kinds of React application tests, and where component tests fit in- A mental model for thinking about the inputs and outputs of the components you test- Options for selecting DOM elements to verify and interact with them- The value of mocks and why they shouldn’t be avoided- The challenges with asynchrony in RTL tests and how to handle them
Prerequisites- Familiarity with building applications with React- Basic experience writing automated tests with Jest or another unit testing framework- You do not need any experience with React Testing Library- Machine setup: Node LTS, Yarn
Master JavaScript Patterns
JSNation 2024JSNation 2024
145 min
Master JavaScript Patterns
Top Content
Featured Workshop
Adrian Hajdin
Adrian Hajdin
During this workshop, participants will review the essential JavaScript patterns that every developer should know. Through hands-on exercises, real-world examples, and interactive discussions, attendees will deepen their understanding of best practices for organizing code, solving common challenges, and designing scalable architectures. By the end of the workshop, participants will gain newfound confidence in their ability to write high-quality JavaScript code that stands the test of time.
Points Covered:
1. Introduction to JavaScript Patterns2. Foundational Patterns3. Object Creation Patterns4. Behavioral Patterns5. Architectural Patterns6. Hands-On Exercises and Case Studies
How It Will Help Developers:
- Gain a deep understanding of JavaScript patterns and their applications in real-world scenarios- Learn best practices for organizing code, solving common challenges, and designing scalable architectures- Enhance problem-solving skills and code readability- Improve collaboration and communication within development teams- Accelerate career growth and opportunities for advancement in the software industry
AI on Demand: Serverless AI
DevOps.js Conf 2024DevOps.js Conf 2024
163 min
AI on Demand: Serverless AI
Top Content
Featured WorkshopFree
Nathan Disidore
Nathan Disidore
In this workshop, we discuss the merits of serverless architecture and how it can be applied to the AI space. We'll explore options around building serverless RAG applications for a more lambda-esque approach to AI. Next, we'll get hands on and build a sample CRUD app that allows you to store information and query it using an LLM with Workers AI, Vectorize, D1, and Cloudflare Workers.