ChatGPT Glitch: OpenAI Fixes Bug Causing Nonsensical Responses in Leading AI Tool

22 Feb 2024 , 03:22 PM

ChatGPT, the generative artificial intelligence tool by OpenAI, experienced a malfunction on Tuesday into Wednesday, producing nonsensical answers to user queries.

OpenAI acknowledged the issue, attributing it to a software tweak that introduced a bug in the language processing of the model. Developers reported ChatGPT generating peculiar responses, including non-existent words, incomplete sentences, and general gobbledygook.

OpenAI identified the cause, implemented a fix, and confirmed the resolution, with ChatGPT returning to normal functionality. Users expressed frustration on the OpenAI discussion forum, describing the experience as if the AI was haunted or compromised.

The incident persisted for over 16 hours before OpenAI updated the status page, assuring that ChatGPT was operating normally.

OpenAI recently secured a deal with investors, reportedly valuing the company at $80 billion or more, marking a significant increase in its value within ten months.

ChatGPT, introduced in late 2022, gained immediate success for its ability to generate text, sounds, and images based on user prompts. OpenAI, known for other AI tools like DALL-E, released a new tool named Sora, capable of creating realistic videos up to a minute long through user prompts.

Microsoft, a major investor in OpenAI with around $13 billion, incorporates OpenAI’s technology into services like Bing and is engaged in AI competition with Google and Amazon.

The US Federal Trade Commission launched an investigation in January into the substantial investments by Microsoft, Google, and Amazon in specialized AI start-ups.

For feedback and suggestions, write to us at [e-mail]