This simple logical question baffles even the most advanced AI

A fascinating new paper from scientists at the AI ​​research nonprofit LAION shows that even the most advanced large language models (LLMs) often face the same simple logical question - a finding that the researchers say casts doubt on whether groundbreaking AI language models are indeed suitable. as advanced as their creators often claim.

The paper, which has yet to be peer-reviewed, refers to the AI ​​stumping prompt as the 'Alice in Wonderland' - or AIW - problem. It's a simple reasoning question: "Alice did that [X] brothers and so do they [Y] sisters. How many sisters does Alice's brother ...