Artificial Intelligence (AI) is the simulation of human intelligence by machine and computer systems. According to Statista, it is expected to represent one of the next great technological shifts -comparable to the invention of the computer. The AI industry is also expected to grow at a constant rate of 154% from 2018 until 2021.
However, with the rapid growth of AI, we must consider the ethical implication that may cause serious problems in the future. Algorithmic bias represents one of these ethical concerns.
Algorithmic bias is defined by TechTarget as being a phenomenon that occurs when an algorithm produces results that are systematically prejudiced due to erroneous assumptions in the machine learning process. It is a common sentiment that AI doesn’t suffer from prejudice or bias due to being driven by mathematical logic, however a large number of examples prove this to be false.
One cause of Algorithmic bias is the reliance on data for deep learning. AI can develop blind spots due to certain data being missing or some being too abundant -leading to bias. A terrible example of this is when Google tagged two black people in a photo as gorillas in a customer’s photos app due to an obvious fault in the algorithm.

Another example found sexist biases in word embedding algorithms causing a tendency for search results to show words such as “engineering” and “programming” to males and “home markers” to females. The repercussions of this could be the industries that are already male dominated to stay as such.
Check out these articles to find out more on algorithmic bias and some shocking examples.













