Is Artificial Intelligence Sexist?
I have a confession to make. By mid-January of this year, I was already suffering from AI fatigue. Does anyone else share that feeling? My inbox was flooded with articles filled with overpromises and unlimited potential for every industry, from financial services, retail, cybersecurity to automotive…. You get the picture.
According to Forrester Research, “the global artificial intelligence (AI) market size was estimated at US$ 119.78 billion in 2022, and it is expected to hit US$ 1,591.03 billion by 2030.” This indicates that AI is here to stay. So I decided to give AI another look and see if AI could benefit my industry and the topics close to my heart: social change, equity, and gender equality.
What better way to test this out than with the rock star of AI, ChatGPT?
My query was simple- “Is ChatGPT sexist?”
Here’s the response:
“No, ChatGPT is not inherently sexist. As an AI language model, I don't hold personal beliefs or attitudes as humans do, and I strive to be unbiased in my responses. However, the data and text that I was trained on may contain biases that exist in our society, including sexism. This means that my responses may sometimes reflect these biases unintentionally. It's important to note that I do not endorse or promote any form of discrimination or prejudice. My goal is to provide helpful and informative responses while being mindful of potential biases and stereotypes.”
Here are my observations.
The speed at which the response was generated was impressive. As a time-strapped 21st-century human, I appreciate any tool that can save me time.
I was impressed that ChatGPT even attempted to answer such a complex question. Most humans shy away from difficult topics.
I was struck that this cutting-edge technology gave me such a mundane and pc answer. Most importantly, I was amused by how its answers basically says: I try my best not to be sexist, but I may have been trained on data that is inherently sexist, so it’s not my fault if I am sexist.
This led me to the next logical question: how can AI technology not be “inherently sexist” but at the same time be “mindful of potential biases and stereotypes?” One could argue that this current iteration of AI has no free will and is completely dependent on the inputs it's given, so it can’t be inherently biased. That’s one area of concern. What happens with the next generation of AI when it does gain consciousness?
My inherent curiosity will push me to continue testing out AI’s potential and its limitations. I’m on the edge of my seat: will AI fulfill its potential as a tool to promote even positively shape social change, or will it just perpetuate some of society’s most ingrained biases, even if it doesn’t mean to?
#socialimpact #equality #AI