OpenAI, the developer behind the renowned ChatGPT language model, has announced an unprecedented postponement of the launch of its new model, GPT-6.1 Astra, initially scheduled for early October. This decision follows significant internal security concerns, as revealed by The Wall Street Journal and confirmed by OpenAI's Chief Security Officer, Saachi Jain, in a statement to CNBC. The model, previously presented as "the smartest and most aligned in the world," is deemed by the company as not "fully meeting the requirements for staying within bounds and capabilities," and "leaves something to be desired" in how it interacts with users regarding the work performed (hln.be).
This synthesis compares media coverage and flags blind spots, biases and conflicts of interest around this topic.
Read the full synthesis