To simplify the full fairness of AI for people with disabilities, we must take small steps
re has been a lot of talk about how artificial intelligence is working to support and even transform the lives of people with disabilities. From specialized applications to robots that act as their aids, technology has accelerated the lives of people with disabilities. While artificial intelligence for people with disabilities has permanently changed its routine, some weaknesses such as discrimination, inequality and prejudice remain in the shadows. To simplify the full fairness of AI for people with disabilities, we need to take a few small steps.
influence of artificial intelligence in people’s lives is no joke. Whether you have a disability or not, AI helps you a lot in your daily life. Furthermore, the whole motto of creating disruptive innovations with the help of artificial intelligence is to automate routine and labor-intensive jobs. One of those time-consuming jobs is finding the right person for a job. It involves ordering the curriculum according to your talents, experience, educational qualifications, abilities, etc. Due to its complex nature, many companies have computerized the resume screening process. most suitable candidates are also invited to participate in the interviews.
While AI takes care of these critical jobs, it omits a part called AI fairness. Yes, initially, discrimination was widely seen on the basis of gender, race and age. But recent reports suggest that discrimination has also reduced job opportunities for people with disabilities. Fortunately, cutting-edge technology companies are taking steps to simplify AI equity for people with disabilities. For example, IBM conducted a workshop at the ASSETS 2019 conference to develop an interdisciplinary community in the areas of AI Fairness, Accountability, Transparency and Ethics (FATE) for the specific situation of people with disabilities. In this article, we walk you through some initiatives companies can come up with to eliminate AI abuse.
Read also This is why Bitwise withdrew the BTC Futures ETF
Spot the downsides
Not all types of disabled people are suitable or unsuitable for an individual. It is based solely on the needs of the job. refore, instead of excluding applications from people with disabilities, the AI should be designed to validate the possibilities of their recruitment. y can first identify the type of disability the person has and compare it to the nature of the job. To do this, the AI application should be fed with unbiased historical data that could identify specific results. Although some disabilities have a direct impact on the nature of work, people can still use technology or software to meet the challenge. For example, if a person is blind and applies for a banking job, we cannot completely rule out the possibilities of their employment. You can use specialized software to deliver your work. AI should be smart enough to see these little dots and pave the way for people with disabilities to have a chance.
Include extreme diversity
Generally speaking, discrimination against people with disabilities can appear very similar to how racial and gender inequalities occur. But this is more than just ignorance. It has many dimensions, varies in intensity and impact, and often changes over time. Furthermore, disability also comes in many forms, including physical barriers, sensory barriers, and communication barriers. refore, to achieve AI equity for people with disabilities, training data sets must be balanced. Technology must be trained equitably so that people with disabilities are not delineated by its developments.
Read also NASDEX is coming to PancakeSwap on November 4
Guarantee data privacy
Like health data, data on people with disabilities is also critical. Many countries refrain from companies requesting data for the disabled due to privacy concerns. This kind of justice through lack of conscience will only make matters worse by hiring an unsuitable candidate for a job. refore, countries should ensure data privacy and loosen their grip on companies’ data collection policies. Doing so will form artificial intelligence carals with unbiased data that value the talents of people with disabilities. Even if the data is anonymized, the normal nature of a person’s situation can make it “re-identifiable.”
Test the bias caral
data reflects the routines of the real world. reason behind discriminatory data is that humanity shows discrimination against certain types of people. While this is not fair, using the same data to perform critical tasks like filtering resumes is not a good thing. refore, AI solution providers must develop a plan to address bias in source data in order to protect themselves from existing discriminatory data. y should also try new ways to increase representation of people with disabilities.
Share this article
Do the sharing thing
Source link
more
