Page 143 - Ai_V1.0_Class9
P. 143

What are the Sources of AI Bias?

                 Some of the sources of AI bias are:
                    • Data: AI systems are the result of the data that is fed into them. The data used to train the AI system is the first
                   step to check for biasness. The dataset for AI systems should be realistic and need to be of a sufficient size.
                   However, the largest data collected from the real world may also reflect human subjectivity and underlying social
                   biases. The Amazon AI recruitment system is a good example. It was found that their recruitment system was
                   not selecting candidates in a gender-neutral way. The machine learning algorithm was based on the number
                   of resumes submitted over a period of 10 years and most of them were men, so it favoured men over women.
                    • Algorithms: The algorithms themselves do not add bias to an AI model, but they can amplify existing biases. Let's
                   look at an example of an image classifier model trained on images in the public domain—pictures of people's
                   kitchens. It so happens that most of the images are of women rather than men. AI algorithms are designed to
                   maximise accuracy. Therefore, an AI algorithm may decide that the people in the kitchen are women, despite
                   some of the images being of men.
                    • Developers: The last source of AI bias is developers. Those who design AI models focus on achieving the
                   desired goals. On that path at times, the biases of the developers are reflected in their models. It's important to
                   note here that ethics and AI bias are not the problems of the machines but of the humans behind the machines.


                 AI Access

                 AI access can be acquired by two means:
                    • Data availability: AI needs access to huge data sets so that it can analyse, draw conclusions, and learn from
                   them.
                    • Abilities: AI needs access to capable hardware to turn its learning into useful action. Self-driven cars are
                   examples of AI with right access.

                 AI is used on bigger, faster and more expensive machines. AI is a privilege that only a few people can afford and
                 take advantage of this new technology. This has created a gap between these two classes of people and it gets
                 widened with the rapid advancement of technology.


                 Difference between Ethics and Morals
                 Ethics  and  morals  are  related  concepts  often  used  interchangeably,  but  they  have  distinct meanings and
                 applications. The word ethics originated from the Greek word ethos. The meaning of ethos is character. The word
                 morals originated from the Latin word mos.
                 The meaning of Mos is custom.


                     Aspect                        Ethics                                      Morals
                  Definition     Rules provided by an external source        Principles regarding right and wrong held by
                                                                             an individual
                  Source         Institutions, organisations, societal norms  Personal beliefs, cultural norms, religious
                                                                             teachings
                  Application    Specific situations and professional practices  Personal behaviour and conduct

                  Objective      Maintain order and fairness in society      Foster personal integrity and align with
                                                                             personal values

                  Examples       Medical ethics, business ethics, legal ethics  Personal beliefs about honesty, integrity,
                                                                             kindness



                                                                         AI Reflection, Project Cycle and Ethics  141
   138   139   140   141   142   143   144   145   146   147   148