A big part of Apple's "reasoning collapse" was models hitting an output limit
The 2025 "Illusion of Thinking" study gets cited to prove reasoning models can't really reason, but its headline Tower of Hanoi failures were largely the models running out of room to print the answer.
OpenAI funded FrontierMath and had the answers before o3 scored 25% on it
Epoch AI built 350 unpublished problems that a script grades with no human in the loop, and the version of o3 that OpenAI shipped solved about 10% of them when Epoch ran the test itself.
AI is getting measurably better at building AI
The sixty-year-old case for a runaway "intelligence explosion" rests on a real, doubling trend and on one loop that no working system has ever been seen to close.
A model builds a fake citation the same way it builds a true one
No part of a chatbot checks whether a reference is real, and the way models are graded rewards a confident invention over an admitted blank.
The Dutch tax office turned a non-Dutch passport into a fraud score
The higher score sent tens of thousands of parents into a fraud dragnet they could not see or appeal.