When we're at the foothills of RSI, and we're about to kick off a period of accelerated AI progress, how will we actually know that the models are aligned?
How will we know AI models are aligned as RSI begins?
AISummary
Dwarkesh Patel asks how researchers could verify that AI models are aligned once recursive self-improvement, or RSI, starts and accelerates AI progress. The post raises the question without offering a method or answer.
Post on XView on X
Source: Dwarkesh Patel · x.comPublished · added here