Author: Vivek Katial; Publisher: Multitudes; Publication Year: N/A. The following article starts by introducing the profound effect that algorithms and, in particular, decisions from algorithms have on our life today. The author goes on to define algorithmic bias as “the ability of algorithms to systematically and repeatedly produce outcomes that benefit one particular group over another.” The article then goes on to state…
Author: Safiya Noble; Publisher: TIME Magazine; Publication Year: 2018. The following article talks about how Google’s search algorithm is biased against the Black woman. It shows and talks about the author’s own identity as a black woman and she talks about how a simple search of black woman would result in pornography of Black girls on Google where as in fact a similar search of say white woman does not yield such…
Author: Thomas Macaulay, Alejandro Saucedo; Publisher: The Next Web; Publication Year: 2022. The following article discusses how almost all artificial intelligence (AI) algorithms are inherently biased as they do exactly what us the programmers tell them to do and we are all inherently biased ourselves. This resource is an interview with Alejandro Saucedo, Chief Scientist at The Institute for Ethical AI, who discusses his main points of what needs to be…
Author: N/A; Publisher: Institute of Chartered Accountants in England and Wales; Publication Year: N/A. The following article talks about how the boards need to lead their businesses on the “right thing to do,” both in terms of the law, their regulatory responsibilities (i.e., treating customers fairly outcomes) and ethics. When shifting towards greater use of technology, banks, insurers, and investment managers need to be pragmatic about the value of…
Author: Sigal Samuel; Publisher: Vox; Publication Year: 2019. The following article talks about the self-driving Uber cars. It points to a very important ethical issue in data science i.e. the issue that self-driving Uber cars so biased that they cannot detect a dark-skinned pedestrian for stopping. This shows that algorithms are inherently biased and that is probably because historically white males in positions of power…
Author: Sibren Isaacman; Publisher: Loyola University – Maryland; Publication Year: 2018. The following article presents a reality that most data scientists will not be able to actually see the affect of their model or algorithm being implemented which makes it difficult to test the outcome from an ethical standpoint. To remedy this he suggests to stay diligent from the start. “After all, it’s our goal to create automated processes that affect…
Author: Selin Akgun, Christine Greenhow; Publisher: AI and Ethics; Publication Year: 2022. The following article discusses the current and proposed uses of artificial intelligence (AI) in promoting educational outcomes for K-12 students. It asserts that, if unchecked, AI algorithms can perpetuate systematic and human biases, especially against groups who are disadvantaged based on factors such as nationality, race, language, or…
Author: Jennifer Priestly; Publisher: SAS Users; Publication Year: 2018. The following article discusses how the data ecosystem is evolving. Data used to be small, structured, and static. Then it became large, unstructured, and in motion. Now it is massive, integrated, and dynamic. The issues related to ethics are much more complex than they used to be. Why do data scientists need to understand ethics? Well, a few people can…
Author: Sarah Thompson; Publisher: Cornell Chronicle; Publication Year: 2022. The following article discusses how a professor in the Department of Statistics and Data Science at Cornell recently began offering a Data Ethics online certificate program which offers 4 2-week courses, in order to focus on larger ethical issues related to algorithms as well as individual choices. This is a educational certificate that could help data science…
Author: Sara Gerke, Timo Minssen, and Glenn Cohen; Publisher: National Library of Medicine; Publication Year: 2020. The following article is dense, but its biggest takeaway is the impact of getting things wrong is extremely high in the healthcare sector. If an algorithm recommends an incorrect or unsafe oncology treatment, it could cost a patient their lives. The practical example of this is simply having bad training data, but it can also take a bias skew. When…