We hope you enjoy today's show. If you do, could vote for us in the People's Choice Podcast Award? It doesn't take long to cast your vote and it would really mean a lot to both Ryan and I. Thanks very much. Cheers!
To vote, please click here.
Artificial Intelligence (AI) is an incredible technology with potential to shape the future of humanity. However, it is fundamentally flawed. Who messes it up? That’s right, human programmers.
The common misconception is that AI is a pure technological brain built from scratch without the risk of human influence. The truth is the program is incapable of performing tasks beyond the boundaries of the specific situations for which the algorithms were designed.
Take the basic linear regression formula: y = mx + b. This simple equation is a program to figure out the output or “Y” by means of the input “B” through the algorithm “mx.” All that an AI is doing can be summarized at a basic level in this equation. It is finding an output from an input using a fundamental algorithm. Since the AI is simply doing the math, the algorithm is the most important part of any AI.
Understanding these principles, it is apparent what really matters in AI construction is the importance put upon the inputs in the coding process. Currently, AI is not capable of making these decisions; only human coders have that power.
In this episode, we talk about today’s AI and its flaws with our guest Broderick Turner, Ph.D., Assistant Professor of Marketing at Virginia Tech., and founder of TRAP LAB (Technology Race and Prejudice LAB). We explore the inherent biases built into programmers’ decisions when making AI’s—and the ridiculousness of a robot uprising with the technology we have today.
So, What’s AI Good for Then?
Many of us perceive AI as mysterious as a magic trick. However, in reality, AI is a series of algorithms. In other words, AI is merely the means of producing an output from an input based upon the algorithm put in place by the programmer. Therefore, AI tends to work best when the inputs are limited.
Here are a few key moments in the discussion:
03:46 Dr. Turner introduces “Weapons of Math Destruction”, the concept that AIs are merely opinions written in code, and what that means for AI. 05:30 We discuss the absurdity of the idea of an AI takeover and why this would be impossible with our current technology. 09:32 Ryan asks Dr. Turner to explain what scenarios AI is good for and we discuss some of the most successful AI products in the modern world. 13:09 Dr. Turner discusses his own private research including how AI is used in conjunction with YouTube thumbnails to determine optimization for view counts for advertisers. 17:12 We discuss how AI is used to determine which thumbnails will be the default based upon data collected through user trials and how AI is effective in this field. We then discuss how AI audits are conducted in these scenarios. 26:28 We discuss the results of the audits conducted by Dr. Turner and what this means about the user groups of the sites tested. 32:40 Dr. Turner discusses the racial and gender biases found in websites like Twitter and how the algorithms the sites use affect them.
Please tell us how we are doing! Complete this short survey.
Customer Experience Information & Resources
LinkedIn recognizes Colin Shaw as one of the 'World's Top 150 Business Influencers.' As a result, he has 290,000 followers of his work. Shaw is Founder and CEO of Beyond Philosophy LLC, which helps organizations unlock growth by discovering customers' hidden, unmet needs that drive value ($). The Financial Times selected Beyond Philosophy as one of the best management consultancies for the last four years in a row. Follow Colin on LinkedIn and Twitter.
Click here to learn more about Professor Ryan Hamilton of Emory University.
Why Customers Buy: As an official "Infl