San Diego-Based AI Expert Discusses Ethical Concerns with Popular ChatGPT Bot

From writing cover letters and job descriptions to wedding vows and love poems, ChatGPT has proven to be able to tackle a wide variety of requests.

But, what is it? NBC 7’s Dana Williams asked ChatGPT to introduce itself, how it works and what it does in 30 seconds or less.

The chatbot responded with, “Hi, I’m ChatGPT, a cutting-edge language AI model developed by OpenAI. I process natural language inputs, generate human-like text responses and assist with tasks such as answering questions, translating languages, generating stories and much more. I work by predicting the next word in a sentence based on the input provided and selecting the most likely response based on my training data.”

According to Anna Marbut, a professor of practice at the University of San Diego’s Applied Artificial Intelligence Master of Science program, the human-like responses are what’s making ChatGPT stand out. It is not the first language model to exist, but it is one of the first to be so conversational and communicate in a way that people are comfortable with.

“It’s a model that’s been trained on billions and billions of examples of human language,” said Marbut. “It has access to all of those things to predict, you know, what is the most likely next thing that it should say.”

Marbut admitted ChatGPT does not come without concerns. ChatGPT lists a few limitations before users enter a request, like “may occasionally generate incorrect information” and “may occasionally produce harmful instructions or biased content.”

“You’ll see if you try to ask it those kind of [harmful] questions, it will usually say, 'I’m just a language model,' you know, ‘I can’t tell you the answer to this,’ or ‘That might be dangerous, I can’t tell you the answer,'” said Marbut.

There has also been concern about ChatGPT being used to complete school assignments, like essays. Marbut said there could be an application for using ChatGPT to write papers, but that it is something we, as a society, need to come up with parameters for.

NBC 7 asked ChatGPT how it would respond to people who are worried that it might be used to help students cheat. It responded, in part, “Artificial intelligence systems like me are powerful tools, but it’s important to understand that our primary purpose is to assist and augment human capabilities, not replace them. Writing papers is a task that requires critical thinking, creativity, and analysis, which are still uniquely human skills.”

Along those same lines Marbut told NBC 7 that she does not think this tool is going to displace the need for people in certain roles, but rather, shift job descriptions to utilize chatbots.

“You could use ChatGPT to be better at your job. You could be more efficient, right?” said Marbut. “I think we still got a long way to go until computers take over.”