Disability and Artificial Intelligence (AI)

Category Topic: AI and Disabilities
Author: Disabled World
Updated/Revised Date: 2024/02/18
Contents: Summary - Definition - Introduction - Main - Subtopics - Publications

Synopsis: A collection of papers that explore the pros and cons of current and future AI applications for the disabled in everyday situations. Not everything about AI is scary; for those with a disability, certain technologies may be used to help make life a little easier. Using AI to make design widely accessible and inclusive is also good for those without a disability.

Defining Artificial intelligence (AI)

Artificial intelligence (AI): Artificial intelligence (AI) is intelligence - perceiving, synthesizing, and inferring information - demonstrated by machines rather than human intelligence. This includes speech recognition and understanding human speech (such as Siri and Alexa), computer vision, translation between (natural) languages, self-driving cars, and many other current and future applications and uses. There are many fields of AI, including 'robotics,' but one of the most commonly known forms is 'machine learning.' This involves a program applying known information to new experiences and 'learning' how to take this historical information and its experiences into account in future actions. Advanced machine learning is often described as 'deep' learning.
Logic Learning Machine (LLM): A machine learning method based on generating intelligible rules. LLM is an efficient implementation of the Switching Neural Network (SNN) paradigm, developed by Marco Muselli, Senior Researcher at the Italian National Research Council CNR-IEIIT in Genoa. LLM has been employed in many different sectors, including the field of medicine (orthopedic patient classification, DNA micro-array analysis, Clinical Decision Support Systems, financial services, and supply chain management.
Large Language Models (LLMs): Artificial intelligence tools that can read, summarize and translate texts and predict future words in a sentence based on knowledge gained from massive datasets, letting them generate sentences similar to how humans talk and write.

Introduction

Artificial intelligence will revolutionize accessibility and inclusion for persons with disabilities. Even though AI is still in its early stages of development, many companies are working on bringing many ideas to reality. Not everything about AI is scary; for those with a disability, certain technologies may be used to help make life a little easier. Large language models like GPT-3 and the soon-to-be-released GPT-4 will be able to reliably pass the Turing test - a Rubicon for machines' ability to imitate human intelligence.

Main Document

Using AI to make design widely accessible and inclusive is also good for those without a disability. The inclusive design considers the needs of all users as a product or service is being developed, from start to finish. With an inclusive or human-centered design, a person with a disability is simply another individual with specific lived experiences. Organizations that design Artificial Intelligence applications for diversity and edge cases, including individuals with disabilities, will create better solutions and experiences for all users.

ChatGPT AI

ChatGPT is a new artificial intelligence (AI) system, known as a large language model (LLM), designed to generate human-like writing by predicting upcoming word sequences. Unlike most chatbots, ChatGPT cannot search the internet. Instead, it generates text using word relationships predicted by its internal processes.

ChatGPT can score at or around the approximately 60 percent passing threshold for the United States Medical Licensing Exam (USMLE), with responses that make coherent, internal sense and contain frequent insights, according to a study published February 9, 2023, in the open-access journal PLOS Digital Health by Tiffany Kung, Victor Tseng, and colleagues at AnsibleHealth.

Continued below image.

Diagram of a human head and brain outlined on an electronics circuit board.

Continued...

AI Discrimination

Artificial Intelligence bias in job hiring and recruiting causes concern as a new form of employment discrimination. Despite its convenience, AI can also be biased based on race, gender, and disability status and can be used in ways that exacerbate systemic employment discrimination.

The U.S. EEOC has released a technical assistance document, The Americans with Disabilities Act and the Use of Software, Algorithms, and Artificial Intelligence to Assess Job Applicants and Employees, focused on preventing discrimination against job seekers and employees with disabilities. The EEOC's technical assistance document is part of its Artificial Intelligence and Algorithmic Fairness Initiative to ensure that the use of software, including artificial intelligence (AI), used in hiring and other employment decisions complies with the federal civil rights laws that the EEOC enforces.

How Generative Ai Models May Distort Human Beliefs

By the American Association for the Advancement of Science - AAAS.

Generative AI models such as ChatGPT, DALL-E, and Midjourney all have features that may distort human beliefs through their transmission of false information and stereotyped biases, according to Celeste Kidd and Abeba Birhane. In this Perspective, they discuss how research on human psychology can explain why generative AI could be particularly powerful in distorting beliefs. The capabilities of generative AI have been exaggerated at this point, they suggest, leading to a belief that these models exceed human capabilities.

People are predisposed to adopt the information of knowledgeable, confident agents like generative AI more rapidly and with greater certainty, the authors note.

Generative AI can create false and biased information that can be spread widely and repeatedly, both factors that predict how deeply that information can be ingrained in people's beliefs. People are most influenceable when they are looking for information and hold more stubbornly to information once it has been received. As much of generative AI is designed for searching and information-providing purposes at the moment, it may be difficult to change the minds of people exposed to false or biased information provided by generative AI, Kidd and Birhane suggest.

"There is an opportunity right now, while this technology is young, to conduct interdisciplinary studies to evaluate these models and measure their impacts on human beliefs and biases before and after exposure to generative models - before these systems are more widely adopted and more deeply embedded into other everyday technologies," they conclude.

AI for Accessibility Grants

Artificial intelligence has yet to be around long enough to tackle every opportunity. Still, Microsoft wants to speed up innovation with a program called "AI for Accessibility," announced earlier this year with $25 million in funding. AI for Accessibility grants support projects that use AI to empower people living with disabilities. Microsoft is looking for individuals or teams passionate about making the world more inclusive and firmly rooted in the communities they intend to benefit. Microsoft wants to invest in ideas developed by or with people with disabilities. AI for Accessibility grant applications are evaluated on their scientific merit, innovative use of AI technology, and potential for scalability.

Imageomics

Imageomics is an emergent scientific field that combines AI machine learning with biology to extract biological information from images of various life forms. It aims to understand the relationship between an organism's phenotype (observable traits) and its genotype (genetic makeup). Imageomics has made stunning progress and is on the verge of major discoveries about life on Earth.