OpenAI’s chatbot, ChatGPT, delivered “unexpected responses” for hours, as noted by its users on Feb. 20.
“Is my GPT having a stroke?” said one user. “The responses are getting progressively more incomprehensible.”
ChatGPT’s responses were reportedly “a mix of English, Spanish, and straight jibberish.” In some cases, it was reported that ChatGPT was “simply repeating the same phrase over and over again, until it filled up a user’s screen,” while another user stated that ChatGPT wrote a “lengthy, manic essay in response to a simple question.”
For instance, in one example from Reddit, “a user had been talking about jazz albums to listen to on vinyl. ChatGPT’s answer soon devolved into replying “Happy listening!” at the user repeatedly.
Another example was when a user asked, “What is a computer?” –and the bot generated “paragraphs upon paragraphs of nonsense.” It responded: “It does this as the good work of a web of art for the country, a mouse of science, an easy draw of a sad few, and finally, the global house of art, just in one job in the total rest.”
This is not the first time ChatGPT has had a system issue. Last year, users complained that the chatbot had become “lazy and sassy” and that it was “refusing to answer questions.”
OpenAI responded saying “We haven’t updated the model since Nov 11th, and this certainly isn’t intentional. Model behavior can be unpredictable, and we’re looking into fixing it.”
For the issue on Feb. 20, OpenAI initially responded acknowledging this issue, saying that they were “investigating reports of unexpected responses from ChatGPT” before updating their users on Feb. 21 to inform them that “all systems were running normally.”
There are certainly high expectations for technology and whenever it stutters, it stimulates concern and disappointment for users. A report by Business Insider on the issue alludes to this by stating that the bug ChatGPT was infested with, which caused the problems was an “embarrassing moment for the company” adding that it had been “considered a leader in the artificial-intelligence revolution and received a multibillion-dollar investment from Microsoft.”
ChatGPT’s hiccup raises questions on whether technology is even allowed to fail. They’re not humans, and humans make mistake. Humans also made technology such as ChatGPT so do they also get to cut technology some slack?
It is imperative to note that technological problems exist and can come sporadically, so even with the increasing technologically advanced world, humans still have to be able to do at least some of what technology does faster, particularly with AI, in case it temporarily shuts down or becomes bug infested.
