The AI-powered ChatGPT bot took the internet by storm when on Wednesday, Artificial Intelligence (AI) research firm OpenAI announced a prototype of a conversation-based AI chatbot.
It is based on the company’s GPT-3.5 technology model that uses deep learning to carry out human-like conversations, understand natural language, and respond in natural language
People are comparing the prototype to Google because its intuitive solution offers the ability to solve complex problems, much like a personal tutor.
“We trained a model called ChatGPT that interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer follow-up questions, admit errors, challenge incorrect premises, and reject inappropriate requests,” OpenAI writes on its announcement page for ChatGPT .
What is ChatGPT?
- Based on GPT-3.5, ChatGPT is a language model that uses deep learning to generate human-like text.
- Unlike the previous model
- ChatGPT is more engaging while GPT-3 only takes text prompts and tries to follow that up with its own generated text
- ChatGPT is much better at generating detailed text and can even generate poems.
- Memory is another unique feature of the attractive feature as it can remember previous comments in the conversation and recount them to the user.
ChatGPT: How to use it?
- The first step is to visit the ChatGPT website and select ChatGPT.
- Next, you need to register after entering basic information.
- Finally, when it’s ready to use, you can enter anything in the text box at the bottom of the screen that you want the chatbot to answer.
Limitations
There are still some acknowledged limitations with bots: Like previous versions of GPT, ChatGPT is still prone to misinformation and disinformation, it can give inaccurate answers to common problems number.
OpenAI understands these obstacles, noting on its announcement blog:
“ChatGPT sometimes writes responses that sound reasonable but are inaccurate or meaningless. Overcoming this problem is challenging, because: (1) during RL training, there is currently no source of truth; (2) training the model to be more conservative causes the model to reject questions that it could answer correctly; and (3) supervised training will fool the model because the ideal answer depends on what the model knows, rather than what the demonstrator knows.”
Currently, OpenAI has opened the bot for evaluation and beta testing purposes. However, API access is expected next year, through which developers will be able to implement ChatGPT into their own software.
: World’s largest telescope ring, Daosheng Solar Radio Telescope: All you need to know
What is Google Bard? GPT Chat Competitors: All the important details here
Categories: Optical Illusion
Source: pagasa.edu.vn