OpenAI has released GPT-4, a large language model (LLM) that marks a shift from model-centric to data-centric artificial intelligence (AI). LLMs boast billions of parameters and are trained on massive amounts of data for self-supervision during pre-training to produce parent models. LLMs are then fine-tuned for a specific task with smaller, labelled datasets. However, there are various ethical concerns associated with LLMs, including bias, privacy and security risks, and lack of interpretability. As the development and deployment of LLMs require responsible AI practices, companies should establish a clear strategy based on principles like fairness and accountability throughout the ML development lifecycle.

source update: GPT-4 and the Next Frontier of Generative AI – Towards AI