The popular ChatGPT conversational AI model’s creator, OpenAI, recently issued an update to increase the chatbot’s accuracy.
Launched by OpenAI in November 2022, ChatGPT is a chatbot. The GPT-3 family of big language models from OpenAI is the foundation upon which it is built, and it is tweaked using both supervised and reinforcement learning methods.
The upgrade has restored the AI’s functionality after a period of inactivity.
This is the second upgrade to ChatGPT since the model’s launch in November and the first one for ChatGPT this year.
Even if ChatGPT might be more precise now, it still only has information as of 2021. While AI’s knowledge is from 2021, it does provide truthful responses!
It’s challenging to identify the situations in which ChatGPT can provide more accurate results. It would be beneficial if OpenAI could include explicit information about future improvements in the release notes.
“That said, be careful when using ChatGPT as a source of information. Although it provides correct answers to many questions, it’s currently not dependable enough to replace Google.” wrote Search Engine General.
A waitlist for the experimental ChatGPT Professional service, which would essentially eliminate the restrictions on the well-liked chatbot for a price, has been made public by OpenAI. With no throttling and as many communications as needed, the AI tool would always be accessible. The startup is seeking input on price from potential participants but has not specified when the trial programme might begin.
Uncertainty exists over ChatGPT Professional’s potential audience. As of last month, the current edition had more than a million members, although it’s unclear how many of those are’serious’ users as opposed to enthusiasts and interested bystanders.