Talk:AI predictions

Skeptical

Gary Marcus predictions

• 2022-03: Deep Learning Is Hitting a Wall: What would it take for artificial intelligence to make real progress?
• 2024-03: Two years later, deep learning is still faced with the same fundamental challenges
• 2024-04: Evidence that LLMs are reaching a point of diminishing returns — and what that might mean

Gary Marcus other

• 2022-05: [Horse rides astronaut: What nearly everyone got wrong about DALL-E & Google’s Imagen, and why when it comes to AI hype, you still can't believe what you read]
  • 2025-10: Pliny the Liberator

Random Essays

• Jascha's Blog
  • 2022-11: Too much efficiency makes everything worse: overfitting and the strong version of Goodhart's law
  • 2023-03: The hot mess theory of AI misalignment: More intelligent agents behave less coherently
  • 2023-09: Brain dump on the diversity of AI risk
  • 2024-02: Neural network training makes beautiful fractals
• Gian
  • 2025-04: Agency is Eating the World