These percentage estimates of AI's proximity to "human reasoning" are misleading abstractions that mask fundamental qualitative differences in how LLMs and huma

These percentage estimates of AI's proximity to "human reasoning" are misleading abstractions that mask fundamental qualitative differences in how LLMs and humans process information.