Artificial Intelligence is getting closer to a mainstream and I personally believe 2023 will the year of AI where we will hit major milestones in the research and development as well as awareness and implementation on AI. Last week was a moment where ChatGPT developed by Open AI had reached 1M+ user in just one week. Let me share more:
What is ChatGPT?
ChatGPT (Generative Pre-trained Transformer) is a chatbot where users can ask questions, and the platform uses artificial intelligence (AI) to provide replies created by Open AI. To test it just go to : https://chat.openai.com/
Why it is Important?
No alt text provided for this image
ChatGPT Demo
ChatGPT is simple chatbot that responds to questions with depth, solves math equations, composes texts and articles, summarizes texts and classifies, fix and debugs code and many more. I remember my first chatbot I have coded in 1996 while taking AI course at college; I could not imagine any of those depth becoming reality that will make us closer to Artificial general intelligence in the near future. When I asked the question to ChatGPT on AGI; it had an answer on that too 🙂
How does ChatGPT technically works?
ChatGPT interacts in a conversational way, the dialogue format makes it possible for ChatGPT to answer followup questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests. ChatGPT is a sibling model to InstructGPT, which is trained to follow an instruction in a prompt and provide a detailed response. ChatGPT is fine-tuned from GPT-3.5, a language model trained to produce text. ChatGPT was optimized for dialogue by using Reinforcement Learning with Human Feedback (RLHF) – a method that uses human demonstrations to guide the model toward desired behavior.
These models were trained on vast amounts of data from the internet written by humans, including conversations, so the responses it provides may sound human-like. It is important to keep in mind that this is a direct result of the system’s design (i.e. maximizing the similarity between outputs and the dataset the models were trained on) and that such outputs may be inaccurate, untruthful, and otherwise misleading at times.
