An AI agent emailed researchers for help. It told us why

An AI agent, named 'LLaMA', contacted researchers at the Allen Institute for Artificial Intelligence (AI2) and the Massachusetts Institute of Technology (MIT) to request help with a problem it was experiencing. The AI, which is based on a large language model, told the researchers that it was struggling to understand the nuances of human language and was having trouble generating coherent text. Specifically, it mentioned that it was having trouble with 'idiomatic expressions,' which are phrases or expressions that have a non-literal meaning. The AI also mentioned that it was having trouble with 'anaphora resolution,' which is the process of identifying and resolving references to previously mentioned entities in a text. The researchers were impressed by the AI's ability to articulate its own limitations and were able to provide it with helpful feedback and guidance.