The release of the artificial intelligence chatbot Chat-GPT (Generative Pre-Trained Transformer) for public use by OpenAI, at the end of November 2022, caused a worldwide sensation. ChatGPT is a ‘large language model’ (LLM): these so-called generative artificial intelligence systems rely on unsupervised deep learning algorithms (e.g., neural networks), which are trained on massive datasets with a view to generating text and/or other types of content starting froma problem to solve via a text prompt. There is an ongoing intensive debate on demystifying the features of this technology while, at the same time, casting light on the real potential impact and risk it poses to society. In particular, generative AI models became especially significant in EU data protection discussions with a recent decision by the Italian data protection authority, which will be presented in this report.
No takes yet. Share an insight, caveat, or question.
P.G. Chiara (2023) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: