Current AI models are not reliably understandable or interpretable, and researchers cannot predict emergent capabilities in advance, which makes 'grown rather than built' metaphor appropriate and underscores deep uncertainty about future development.
factualpending
Speaker
Sarah Hastings WoodhouseEvidence Quote
“it's a reminder of how sort of uninterpretable they are... we're just kind of like throwing all of this compute and data into these training runs and like seeing what happens”
Created: 8/11/2026, 6:36:18 AM
Your notes
Loading notes...