OpenAI announces ChatGPT chatbot: What is it, how it works, and limitations

ChatGPT is an AI chatbot based on GPT-3.5 series of natural language model, which can understand human language and responding in human-like language

Artificial intelligence (AI) research organisation OpenAI recently announced ChatGPT, an AI chatbot based on the GPT-3.5 series of language models. Early users have described the technology as an alternative to Google, since it can provide descriptions, solutions to complex questions, and more.

"We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer follow-up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests," OpenAI said in a blogpost.

ChatGPT: What is it

ChatGPT is a prototype dialogue-based AI chatbot capable of understanding natural human language and responding in human-like language. The dialogue format makes it possible for ChatGPT to answer follow-up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests. It can also remember earlier comments in a conversation and recount them to the user.

OpenAI has opened up the ChatGPT for users’ feedback and to learn about its strengths and weaknesses. However, the API access is expected to come next year, through which developers will be able to implement ChatGPT into their own software. During the research preview, usage of ChatGPT is free.

Even in the testing phase, people are marveling at its intelligence and abilities.

"ChatGPT could be a good debugging companion; it not only explains the bug but fixes it and explains the fix" tweeted software start-up founder Amjad Masad, who got ChatGPT to spot errors in his code.

ChatGPT: How it works

OpenAI said in a statement that the chatbot was created with a focus on ease of use. Trained by AI and machine learning, the tech is designed to provide information and answer questions through a dialogue-based interface similar to using instant messaging platforms. In its case, however, there is no real person on the other end but an AI chatbot trained on a huge sample taken from the internet.

ChatGPT: Limitations

ChatGPT is currently a prototype. It sometimes writes plausible-sounding but incorrect or factually incorrect answers. Besides, ChatGPT often excessively verbose and overuses certain phrases, such as restating that it is a language model trained by OpenAI. Sometimes, it responds to harmful instructions, inappropriate requests or exhibits biased behavior.

Acknowledging the limitations in its current form, OpenAI CEO Sam Altman wrote on Twitter, “…a lot of what people assume is us censoring ChatGPT is in fact us trying to stop it from making up random facts. Tricky to get the balance right with the current state of the tech. It will get better over time, and we will use your feedback to improve it.”