The BasePeople

Fazl Barez

Titles as the sources gave them. In 1 entry, with 2 checked quotations.

What they said

Newest first. Each quotation was found word for word in the source.

“Yet we continue to build increasingly powerful systems without fully understanding how they work, how they will behave in unfamiliar situations, or how to reliably control them.”
Understanding AI from the Inside: The State of Mechanistic Interpretability for AI SafetyarXiv, November 5, 2026. Speaking as: Other. Motivation in the abstract of the survey, which the authors present as written jointly.
“We survey mechanistic interpretability for AI safety in order to answer the question: Does understanding how models work actually make them safer?”
Understanding AI from the Inside: The State of Mechanistic Interpretability for AI SafetyarXiv, November 5, 2026. Speaking as: Other. The abstract's statement of the survey's central question.

In the Base

Every entry they appear in.

  1. ResearcharXivWrote it

    Understanding AI from the Inside: The State of Mechanistic Interpretability for AI Safety

    According to the abstract, Barez and 26 co-authors survey mechanistic interpretability and ask whether understanding how models work makes them safer. They propose a framework for evaluating interpretability claims, discuss limits, and release code notebooks.