Mechanistic Interpretability · EP 8 · Aug 18, 2026

How Do AI Brains Actually Work? The Truth

Name: How Do AI Brains Actually Work? The Truth
Uploaded: 2026-08-18T20:36:13.044Z
Description: The newest research in Mechanistic Interpretability, explained.

This week in Mechanistic Interpretability

Papers covered

Circuit findings fall apart under scrutinyarXiv:2608.13754
Training dynamics obey measurable lawsarXiv:2607.27281
Routing and answers are separate systemsarXiv:2608.11767
Robot AI tracks task progress internallyarXiv:2608.13474
AI agent automates interpretability researcharXiv:2608.12036

Your field, every week

Get this for your own field.

Sciport turns the newest papers in your corner of the literature into a short weekly video like this one — auto-curated, delivered to your inbox. First episode’s on us.

Generate your first video →