Don't Just Read the News, Understand It.
Published loading...Updated

Apple Study Reveals AI Reasoning Models Collapse on Complex Problems

  • On June 6, Apple published a paper titled 'The Illusion of Thinking' revealing that large reasoning models fail at complex logic tasks like Tower of Hanoi.
  • The research arose from testing reasoning-optimized models on puzzles and benchmarks, showing their accuracy collapses as complexity increases despite access to correct algorithms.
  • Apple found that models generate hallucinations up to 48% of the time and lack generalizable problem-solving skills, with performance dropping to zero beyond certain complexity thresholds.
  • The authors found that model performance sharply declines and eventually stops altogether once problems surpass a specific complexity level, indicating that current models depend more on pattern recognition than on genuine reasoning.
  • These findings challenge claims about near-term artificial general intelligence and imply fundamental limits in large reasoning models that urge more rigorous scientific analysis.
Insights by Ground AI
Does this summary seem wrong?
Podcasts & Opinions

65 Articles

All
Left
9
Center
9
Right
3
Lean Left

A paper from employees of the Group raises doubts that AI will achieve a "thinking ability" comparable to humans in the foreseeable future

·Vienna, Austria
Read Full Article
Lean Left

Apple’s study reveals that some of the most sophisticated models fail shockingly when faced with difficult logical tasks, leaving open the question about the real potential of AI

·Buenos Aires, Argentina
Read Full Article
Think freely.Subscribe and get full access to Ground NewsSubscriptions start at $9.99/yearSubscribe

Bias Distribution

  • 43% of the sources lean Left, 43% of the sources are Center
43% Center
Factuality

To view factuality data please Upgrade to Premium

Ownership

To view ownership data please Upgrade to Vantage

Hindustan Times broke the news in New Delhi, India on Saturday, June 7, 2025.
Sources are mostly out of (0)