How ChatGPT Might Not Be What It’s Cracked Up To Be
Chatbots have been around for ages. With rumors flying that they soon may take over even more jobs than ever before some researchers are scrambling to explain what this means. Our world is ever-changing and as some technologies are continually advancing we, as humans, are going to have to work with AI or find a new job.
What is ChatGPT?
Developed by OpenAI, a California-based a 2015 startup, with early funding from Elon Musk, ChatGPT can write essays and program code within seconds. After investing $1 billion in OpenAI in 2019, Microsoft just signed a multi-billion dollar deal with the firm. Most recently, this artificial intelligence took a medical exam and scored passing or near passing grades. Which has people wondering if ChatGPT will be replacing humans in the workforce and widen wealth inequality.
The Rumors
While past advancements in AI have affected factory jobs, this time it’s coming into the office. So, what white-collar jobs are we going to see most affected?
- Finance jobs such as financial analysts and personal financial advisors.
- Tech jobs such as coders, computer programmers, software engineers and data analysts.
- Legal jobs such as legal assistants and paralegals.
- Media jobs such as advertising, journalism, content creation and technical writing.
But could AI actually replace human workers? According to The Guardian, “Workforce experts are asking many questions about AI: will it displace many call-center workers or merely make those workers more productive by quickly delivering needed information to them as they speak to customers? Will AI displace radiologists because of its ability to read cancer scans or will it help radiologists by enabling them to focus on more complicated, nuanced issues in interpreting scans? Will AI be able to replace some journalists by writing stories about baseball games or Wall Street’s daily ups and downs?”
The Problem with ChatGPT
There is a massive flaw with OpenAI’s chatbot and we are sure some people saw it coming from a mile away. Certain clusters of words can break it! If you’ve ever used an AI program to help you write a term paper or dissertation, then you know what it’s doing–it simply does not contain human syntax. The more they advance the convincing, machine-learning ChatGPT, the more strange things it does. According to Vice, “When an earlier model was asked to repeat “StreamerBot,” for example, it said, “You’re a jerk.”
Disinformation researchers have even speculated that ChatGPT would become the next tool for spreading fake news like wildfire. And, while most businesses won’t be using it to decipher a Reddit thread or game on Steam, there are over 100 phrases that the chatbot cannot describe correctly. Since ChatGPT does have limitations, it probably won’t be taking over the copywriting or advertising industry any time soon, because it simply isn’t superior to the human mind.
At PCX quality is our focal point–we offer it in our service, semiconductors, and even our news.