Watermarks act at the model’s moments of doubt, and so do the safety checks that catch AI mistakes
Javier Marín Valenzuela
The number that fooled every hallucination detector
It’s a feature of the architecture
How geometry shows when LLMs are lying
A story about failing forward, spheres you can’t visualize, and why sometimes the math knows things before we do
What high-resolution NN training dynamics taught me about feature formation
How selective amplification emerged across evolution, chemistry, and AI through convergent mathematical solutions
Here’s why it happens — and how to fix it
Towards new forms of artificial moral agency