Miles Brundage calls 'recursive self-improvement' AI framing vague and misleading
Overview
Miles Brundage, a former OpenAI policy researcher, says on X that the phrase "Recursive self-improvement (and we'll figure out later what role humans play)" has the same problem as Tesla's "Full Self-Driving (Supervised)".
He argues both labels are vague enough to satisfy critics while misleading about what the goal actually is. The post is his only report so far; it names no specific company or AI lab as the source of the phrase.
Written by AI from the articles below · updated Oct 9, 6:07 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
Miles Brundage@Miles_BrundageXBrundage says recursive self-improvement claims are vague and misleadingAIMiles Brundage compares the phrase "Recursive self-improvement (and we'll figure out later what role humans play)" to Tesla's "Full Self-Driving (Supervised)". He says both are vague enough to satisfy critics while misleading about the actual goal.
Heat trend
Not enough continuous observations to show a trend yet.