The AI-powered ChatGPT bot took the Internet by storm when Artificial Intelligence (AI) research company OpenAI announced the prototype of the dialogue-based AI chatbot on Wednesday.
It is based on the company’s GPT-3.5 technology model that uses deep learning to engage in human-like conversations, understand natural language, and respond in natural language.
People are comparing the prototype to Google, due to its on-the-sight solution that provides capabilities for complex problems, like a personal tutor.
“We have trained a model called ChatGPT that interacts conversationally. The dialogue format makes it possible for ChatGPT to answer follow-up questions, admit its errors, question incorrect premises, and reject inappropriate requests,” OpenAI wrote on its announcement page for ChatGPT.
What is ChatGPT?
- Based on GPT-3.5, ChatGPT is a language model that uses deep learning to produce human-like text.
- Unlike the previous model
- ChatGPT is more engaging, while GPT-3 just took text prompts and tried to continue with its own generated text.
- ChatGPT is much better at generating detailed text and can even create poems.
- Memory is another unique feature of hot as it can remember previous comments in a conversation and tell them back 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 the basic details.
- Now, finally, when it’s ready to use, you can type anything in the text box at the bottom of the screen that you want the chatbot to respond to.
the limitations
There are still some limitations with the bot that were acknowledged: like previous versions of GPT, ChatGPT is still prone to misinformation and bias, it can give incorrect answers to, for example, algebraic problems.
OpenAI understands these setbacks and noted in its announcement blog:
“ChatGPT sometimes writes answers that sound plausible but incorrect or nonsensical. Solving this problem is challenging, since: (1) during RL training, there is currently no source of truth; (2) training the model to be more cautious causes it to reject questions it can answer correctly; and (3) supervised training confuses the model because the ideal response depends on what the model knows, rather than what the human demonstrator knows.”
For now, OpenAI has opened up the bot for evaluation and beta testing purposes, although API access is expected to follow next year with which developers will be able to implement ChatGPT in their own software.
READ ALSO: World’s largest telescope ring, Daocheng Solar Radio Telescope: everything you need to know
What is Google Bardo? GPT Chat Rival: All important details here
Categories: Optical Illusion
Source: ptivs2.edu.vn