JUST IN: A new benchmark challenges the narrative that artificial general intelligence is near.

The assessment suggests that, despite recent advances, current systems still exhibit significant limitations when subjected to more demanding tests.
A new benchmark casts doubt on whether AGI is an imminent reality.
The measurement suggests that current models still fall short in capabilities that would require broader and more consistent reasoning.