On social media, reports are screaming of a chatbot owned by OpenAI.
Due to an embarrassing bug, the highly anticipated artificially intelligent (AI) tool ChatGPT has confused users by responding to their queries in Spanglish.
Social media users have been posting reports about an OpenAI-owned bot speaking incoherently, and the US tech company has acknowledged that users are getting “unexpected responses.”
Words composed of a mix of Spanish and English were among the odd replies, and some people kept repeating the same word.
As one example put it: Please let me know if there are any more amazing things your articulation is craving!
According to AI researcher and New York University professor Gary Marcus, ChatGPT has “gone insane.”
Mr. Marcus stated in his newsletter that “these technologies weren’t ever reliable.”
There is no way that safety guarantees can be engineered around them. The alchemy of machine learning is still in its infancy.
A user published on the organization’s panel that, As of roughly three hours ago, all of my interactions with GPT4 degraded very rapidly into crap.
Many users said that ChatGPT would function generally until the replies got absurd.
ChatGPT stated in one post: A day. a peak. A chalk. An igniter. An outline. A ken. A vire. A vane. a byre. Goodbye. A had—a refusal.
While acknowledging that a few users had encountered difficulties, OpenAI stated that the problem had been located and was being fixed.
Since then, the issue has been resolved.
ChatGPT, which can respond to complex inquiries with responses that sound human, has been called a significant breakthrough in AI technology.
Companies are utilizing bots to automate repetitive work. They might use them to compose emails, compile reports, or respond to client inquiries.
Technology giants have invested millions of dollars in artificial Intelligence startups to create “large language models,” which are trained on massive databases containing millions of articles and countless books.
While Google and Amazon have spent millions of dollars in competitor Anthropic, Microsoft has committed to investing $13 billion in OpenAI alone.
Elon Musk is also establishing a competing business of chatGPT named xAI.
But critics have long noted that chatbots can sometimes respond strangely, especially since they might fabricate information, a phenomenon referred to as “imagining things.”
This week, OpenAI revealed a new artificial intelligence tool that can create lifelike videos using text lines.
On the other hand, some of its videos include strange distortions in which people, animals, or things emerge out of thin air.
The CEO of OpenAI, Sam Altman, has said that AI could be a fatal threat to humankind.
He has been gathering millions of dollars simultaneously to research and develop ever-more-powerful variants of the technology.
The Government’s Artificial Intelligence Security Summit in November was partially motivated by worries about the unknown threats posed by AI. It resulted in an agreement to establish a safety center to monitor the threat. The United States government has been testing ChatGPT on its website.
