Dear Colleagues! This is Asrar Qureshi’s Blog Post #801 for Pharma Veterans. Pharma Veterans welcome sharing of knowledge and wisdom by Veterans for the benefit of Community at large. Pharma Veterans Blog is published by Asrar Qureshi on WordPress, the top blog site. Please email to firstname.lastname@example.org for publishing your contributions here.
Quite recently, the world has been abuzz with Chat GPT, first GPT3 and then more powerful GPT4. The onslaught of talks, comments, ideas, forecasts, and possibilities is so fierce that it looks like everything will be displaced by the category 5 cyclone of GPT; of course, more versions shall keep coming.
It is being forecasted that many, if not most, jobs will be lost to GPT. Since GPT4 is also self-learning, so its capability shall keep enhancing with time. Even now, experiments have shown its equivalence to human intellect in certain pursuits; for example, it defeated a chess master in a game, or it passed MBA of Stanford.
We Who Do Not Know Chat GPT – are the majority in Pakistan, even among educated classes. I shall not go into the reasons, just stated the fact. So, I thought to understand the basics about GPT and share with you.
Let us start from the basics!
GPT – stands for Generative Pre-trained Transformer. It is among the most remarkable creations of Artificial Intelligence; AI is already storming the world. Artificial Intelligence as defined by John McCarthy in 2004 “It is the science and engineering of making intelligent machine, especially intelligent computer programs. It is related to similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.” The concept was first triggered by Alan Turing, considered the father of computer science. AI works the same way as we do. For doing anything, we refer mentally to our learnings, knowledge, and skills, recall the data in milliseconds and take action. AI was doing exactly the same, with the difference that it was limited to what had been fed to it. GPT has changed this basic premise; it started from huge data fed into it, and it is now enhancing its learning as it is exposed to more new data. In this sense, it is more alive.
ChatGPT combines the chat function with GPT through chatbot technology. It is an AI language model developed by OpenAI, and since its introduction, ChatGPT has captured the imagination of many with its ability to generate human-like text and engage in intelligent conversations.
Trained on vast amounts of data, including books, articles, and websites, it can comprehend and generate text across various topics. Chat GPT has proven capable of simulating conversations, offering insightful responses, and providing valuable information to users.
Chat GPT Applications – Few Examples
The applications of Chat GPT are vast and varied. It can serve as a powerful tool for content creation, generating high-quality written material for articles, blogs, and social media posts. Chat GPT’s conversational abilities make it an ideal candidate for virtual assistants, customer support systems, and chatbots, improving user experiences and streamlining interactions.
Moreover, Chat GPT can enhance language translation and interpretation services, enabling more accurate and natural language translations in real-time. It can assist in personalized recommendations, helping users discover relevant products, services, or content tailored to their preferences.
In the field of education, Chat GPT can act as a virtual tutor, providing personalized learning experiences, answering questions, and explaining complex concepts. It also has the potential to revolutionize research and information retrieval, aiding researchers in analyzing vast amounts of data and summarizing research papers efficiently.
Some Myths and Misconceptions
Despite its advancements and potential, Chat GPT has faced misconceptions. One common concern is that AI language models like Chat GPT will replace human jobs. However, the primary purpose of these models is to augment human capabilities, assisting in tasks and enhancing productivity rather than replacing human interactions.
Another misconception is that AI models generate information that is always accurate. It is crucial to recognize that Chat GPT’s responses are based on patterns learned from training data, and it may generate plausible sounding but factually incorrect information. Careful verification and critical thinking are essential when utilizing AI-generated content.
A more serious misconception is that AI enabled machines shall become so powerful that these will take over humans. Movies like Matrix and many others reinforce this concept. There is no doubt that the power of AI must be harnessed to keep it useful. Very recently, CEO of OpenAI, Sam Altman has presented himself before the American Congress and showed his concern about misuse of AI, which may result in various forms, weapons, means of communication etc.
The development and deployment of AI language models like Chat GPT raise important ethical considerations. Ensuring transparency, privacy, and accountability is crucial. Users should be aware when interacting with an AI system and understand the limitations of AI-generated content.
Guarding against biases is another key aspect. Training data for AI models can inadvertently reflect societal biases present in the source material. Efforts should be made to mitigate these biases and promote fairness and inclusivity in AI systems.
As Chat GPT continues to make strides in natural language processing, it is important for those who are unfamiliar with this technology to gain a better understanding of its potential applications, limitations, and ethical considerations. While Chat GPT offers exciting possibilities, it is vital to approach its usage with caution, verifying information and recognizing the role of human oversight in ensuring responsible and beneficial AI deployment. By embracing AI language models like Chat GPT, we can unlock new realms of productivity, creativity, and knowledge-sharing in the digital age.
Disclaimer: Most pictures in these blogs are taken from Google Images and Pexels. Credit is given where known; some do not show copyright ownership. However, if a claim is lodged at any stage, we shall either mention the ownership clearly, or remove the picture with suitable regrets.