ChatGBT boss worries about misuse of AI technology

Article content

A 37-year-old tech guru admits he’s a little afraid of his own company’s invention — ChatGPT.

Article content

OpenAI CEO Sam Altman spoke to ABC News about the notorious AI chatbot and said he felt care must be taken with this technology because of its power.

Article content

“I think people should be happy that we are a little bit scared of this,” he said, noting the technology, while potentially dangerous, could also be “the greatest technology humanity has yet developed” to improve lives.

Good Morning America reports that Altman’s interview focused on the rollout of GPT-4, the latest AI language model.

(ChatGPT is an Artificial Intelligence language model; GPT stands for Generative Pre-trained Transformer.)

It is the fastest growing consumer app, having achieved 100 million monthly users over a few months.

GPT-4 scored in the 90th percentile on the Uniform Bar Exam. It also got an almost perfect score on the SAT Math test and can proficiently write computer code in most programming languages.

Article content

RECOMMENDED VIEO

Altman said in the interview that he is concerned with these models being used to generate disinformation; their capacity to write computer code also has the potential for their use in cyberattacks.

Generally speaking, Altman’s concern is not that the machine will think for itself or take over humanity, but that the wrong people will be in control of this type of technology.

“There will be other people who don’t put some of the safety limits that we put on,” he said. “Society has a limited amount of time to figure out how to react to that, how to regulate that, how to handle it.”

The hope is that the model will learn to become a reasoning engine over time, able to separate fact from fiction.

RECOMMENDED VIEO

Currently, there is an issue with the program giving users inaccurate information.

There are concerns this technology will eventually replace jobs. Students are already using ChatGPT to write their assignments for them.

Fortunately, a Princeton student — from Toronto — has already created an app that can tell if ChatGPT wrote an essay.

NPR reports that GPTZero, the creation of Canadian Edward Tian, can quickly ascertain whether a person or AI wrote an essay.

Tian is a computer science major who is minoring in journalism.