Page 333 - Artificial Intellegence_v2.0_Class_11
P. 333

At a Glance


                       •  Artificial intelligence ethics is a set of moral principles and practices designed to support the development
                       and responsible use of artificial intelligence technology.
                       •  For any company, AI Ethics framework is important because it highlights the risks and benefits of artificial
                       intelligence tools, and provides guidelines for their responsible use.
                       •  AI has huge potential to do good. AI is helping people in the fields of medicine, cyber security, fraud detection,
                       agriculture, gaming etc.
                       •  UNI Global Union has specified 10 key principles for Ethical AI.
                       •  The key principles include transparency, relegating machines to tools, sharing the benefits of AI, and having
                       a human in control approach among many others.
                       •  Bias is any prejudice against individuals or groups, especially in ways that are considered unfair.
                       •  Bias in AI is a term used to describe situations where ML-based data analysis systems are biased against
                       certain groups of people. These prejudices usually reflect the prevailing social prejudices about race, gender,
                       biological sex, age, and culture.
                       •  Though bias can and will enter AI systems, steps can be taken to reduce bias in AI.


                                                              Exercise





                                                         Solved Questions


                                                  SECTION A (Objective Type Questions)
                          uiz

                 A.   Tick ( ) the correct option.
                     1.   Ethical AI governance bodies on global and regional levels all __________________ .
                          a.  Stakeholders                               b.  Companies
                          c.   AI project cycle                          d.   Neural Networks

                     2.   Data bias means that the __________________ are trained using biased data.
                          a.  prejudice                                  b.   humans
                          c.   algorithms                                d.   Either b or c

                     3.   One way to reduce bias is:
                          a.  The programmer should be held responsible for any bias.
                          b.  To make sure you choose a representative dataset.
                          c.  Companies should invest more, and take a multidisciplinary approach to bias investigations.
                          d.  Both b and c

                     4.   Any __________________ can transfer from the programmer to the AI machine while the algorithm is being developed.
                          a.  Ethics                                     b.   Pattern
                          c.   Bias                                      d.   Data

                     5.   AI  Ethics  framework  is  important  because  it  highlights  the  __________________  and  __________________  of  artificial
                          intelligence tools.
                          a.  risks, benefits                            b.   bias, pattern
                          c.   pattern, data                             d.   prejudice, bias

                                                                                   AI Values (Bias Awareness)      331
   328   329   330   331   332   333   334   335   336   337   338