AI: misconceptions and myths
Artificial intelligence has been the biggest trend in startups in the last years, it has gotten into a point that AI is no longer a differentiator between competitors. Carl Eschenbach from Sequoia Capital even joked in his talk at our class that whenever they listen to a pitch that does not mention AI or Machine Learning they think there is something odd going on.
AI in startups came to the point that makes you wonder, are they really using AI? Or are them confusing AI with something else? And not only this is happening, but taking a look at the public: their reaction varies from very interested on this new technology or scared about it. As it usually happens with big complicated topics there are many myths and misconceptions about it, in this article I would like to explore them.
To understand what has been going on, let’s start with the basics, what is AI?
There are many definitions for AI, and maybe that is a source of the problem of mistaking AI with something different. Professor John McCarthy from Stanford, also known as the father of AI, says that AI is ‘ is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.’ (http://jmc.stanford.edu/artificial-intelligence/what-is-ai/index.html)
AI, ML and DL are the same thing
It seems like a common mistake where people think the Artificial Intelligence, Machine Learning, Deep Learning are synonyms. This assumption is understandable as it is not easy to comprehend everything that AI involves.
Nevertheless we can differentiate them: Deep Learning is part of Machine Learning and Machine Learning is part of Artificial Intelligence. This means that DL solves a problem from ML and ML learns from data and in consequence takes care of an specific AI issue. (https://medium.com/luminovo/mythbusters-7-common-myths-in-ai-98c26d9b3e4a)
AI will take over every job
There is an arguably irrational fear that AI will take over human jobs. This emotion is tightly related with the fear of change, meaning that people are at first not open for it and it takes a longer time to adapt. Nevertheless, there is some true in this statement: AI is transforming the workplace. I choose to see it as a positive thing as AI is helping us solve problems and gives us time to focus on other problems.
But do not be afraid, AI is creating new jobs, we just do not know what they are yet. There is a report that 85% of the jobs of the future do not exist yet. (https://www.huffingtonpost.ca/2017/07/14/85-of-jobs-that-will-exist-in-2030-haven-t-been-invented-yet-d_a_23030098/)
There are many more myths that we could look into, for me these ones were the most important ones. How about you, would you agree? Which one do you think is the biggest or most common misconception of AI?
One comment on “AI: misconceptions and myths”
Comments are closed.
Great read. Another thing that I noticed when people talk about AI is how many think that it is a new concept and is a recently developing field. The truth of the matter is that scientists and engineers have been working on AI based systems for a long time and although interest in the field, be it academic or commercial, has been on an incline in recent years, AI has maintained a steady interest in the minds of the relevant people in the field and is not something which, in my opinion, should be viewed as a recently advancing area.