What is CHATGPT ?

If you have been connected with the IT news over the last month, you have undoubtedly heard about ChatGPT — the new AI chatbot from OpenAI. As Andrew Ng rightly said, AI is the new electricity. It is set to revolutionize each aspect of our life, and ChatGPT will change the entire Software Development Life Cycle. It might just seal the fate of some developers. Application development speeds will zoom, and costs will fall drastically.

The current beta version is free for all — to invite more audience. With more and more enthusiasts trying it out, ChatGPT has created a new wave in the IT world. Some have already made millions out of it, and others are on the verge of losing their jobs. The revolution is knocking on your door. Embrace it, or get swept away — the choice is yours.

Basics

Okay, so what exactly is this OpenAI or ChatGPT? OpenAI is an AI research and deployment company that owns and exports several APIs that access their AI models, so we need not invest in developing and reinventing complex AI models. Instead, we can use their API (as a SAAS) and build our applications.

A few weeks back, they released the latest chat API, the ChatGPT, which has impressive capabilities. “Impressive” is a gentle word. However, if you try to chat with the ChatGPT, you will indeed have many more expletives in your mind.We trained this model using Reinforcement Learning from Human Feedback (RLHF), using the same methods as?InstructGPT, but with slight differences in the data collection setup. 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. We mixed this new dialogue dataset with the InstructGPT dataset, which we transformed into a dialogue?format.

To create a reward model for reinforcement learning, we needed to collect comparison data, which consisted of two or more model responses ranked by quality. To collect this data, we took conversations that AI trainers had with the chatbot. We randomly selected a model-written message, sampled several alternative completions, and had AI trainers rank them. Using these reward models, we can fine-tune the model using?Proximal Policy Optimization. We performed several iterations of this?process.

ChatGPT is fine-tuned from a model in the GPT-3.5 series, which finished training in early 2022. ChatGPT and GPT 3.5 were trained on an Azure AI supercomputing?infrastructure.

Limitations

• ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers. Fixing this issue is challenging, as: (1) during RL training, there’s currently no source of truth; (2) training the model to be more cautious causes it to decline questions that it can answer correctly; and (3) supervised training misleads the model because the ideal answer?depends on what the model knows, rather than what the human demonstrator?knows.
• ChatGPT is sensitive to tweaks to the input phrasing or attempting the same prompt multiple times. For example, given one phrasing of a question, the model can claim to not know the answer, but given a slight rephrase, can answer?correctly.
• The model is often excessively verbose and overuses certain phrases, such as restating that it’s a language model trained by OpenAI. These issues arise from biases in the training data (trainers prefer longer answers that look more comprehensive) and well-known over-optimization issues.
• Ideally, the model would ask clarifying questions when the user provided an ambiguous query. Instead, our current models usually guess what the user?intended.
• While we’ve made efforts to make the model refuse inappropriate requests, it will sometimes respond to harmful instructions or exhibit biased behavior. We’re using the?Moderation API?to warn or block certain types of unsafe content, but we expect it to have some false negatives and positives for now. We’re eager to collect user feedback to aid our ongoing work to improve this?system.

Iterative deployment

Today’s research release of ChatGPT is the latest step in OpenAI’s?iterative deployment?of increasingly safe and useful AI systems. Many lessons from deployment of earlier models like GPT-3 and Codex have informed the safety mitigations in place for this release, including substantial reductions in harmful and untruthful outputs achieved by the use of reinforcement learning from human feedback?(RLHF).