What is OpenAI’s ChatGPT, how it works, and can it replace humans?
The world of technology is buzzing with a new word – ChatGPT. The platform was made available for public beta testing on December 1. Within a week of its launch, ChatGPT has crossed more than 1 million users. There are speculations that the platform can replace Google search and more so humans in the future. Amazed and want to know more about it? Well, you have come to the right place. We explain to you everything about ChatGPT – what it is, how it works, and more.
What is ChatGPT?
ChatGPT is a conversational chatbot. It has been developed by Elon Musk-founded independent research body OpenAI. The organisation was co-founded by its current CEO Sam Altman and Elon Musk in 2015.
Coming back to ChatGPT, it is trained to follow instructions in a prompt and provide a detailed response. Users can simply feed in their query and the chatbot will reply to them. But then how is it different from other Artificial Intelligence (AI) chatbots? As per the creators, ChatGPT, unlike other AI chatbots, can answer follow-up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests.
How does ChatGPT work?
As per OpenAI, ChatGPT is a free service, but only during the research preview. This means one can simply head to the OpenAI website and click on the Try ChatGPT button to begin using the platform. You can either sign up or use your OpenAI account to start using ChatGPT. The company has also given a sample on the web page for reference.
OpenAI says that it has trained this model using Reinforcement Learning from Human Feedback (RLHF) to make it more fine-tuned and conversational. “We trained an initial model using supervised fine-tuning: human AI trainers provided conversations in which they played both sides—the user and an AI assistant. We gave the trainers access to model-written suggestions to help them compose their responses”, it says.
As mentioned before, ChatGPT is free to use for the research period only. OpenAI CEO Sam Altman has already hinted that the company may monetize the platform in the future. “We will have to monetize it somehow at some point; the compute costs are eye-watering”, he said in a tweet when asked if the service will forever be free.
Are there any limitations of ChatGPT?
ChatGPT is the latest technology in the Generative Pre-Trained Transformer (GPT) family. To put in simple words, it is the latest tool in auto text-generating AIs. But, it is not free from errors or limitations. OpenAI, on its website, admits that ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers. Also, the model is often excessively verbose and overuses certain phrases. Also, the chatbot is sensitive to how the input is phrased. For example, it may have an answer to a query phrased in one style, but the model may not know the answer if given a slightly different phrase.
Can ChatGPT replace Google search or humans?
We all excessively use Google to find answers to our daily queries. Can we rely on ChatGPT for the same? Now, despite the model being the latest generation of technology, it is not without flaws. Professionals are pointing out the platform’s shortcomings. For example, Steven T Piantadosi, a professor at UC Berkeley has shown in a series of tweets how ChatGPT is not free from biases and that its filters can easily be bypassed with simple tricks. So the answer to the question above – is ChatGPT capable of replacing Google search or humans – is a simple no. At least for now.
Start your day, the smart way!
1 Mail | 2 Mins | 3 Stories