Study Guide: Chapter 11 — Multipolar Scenarios¶

Core Idea¶

A multipolar (many competing superintelligences, no singleton) outcome isn't automatically safe. Cheap machine labor could crash human wages below subsistence (the horse precedent); capital ownership and political redistribution could offset this short-term, but the Malthusian principle likely reasserts over the long run as digital populations reproduce faster than resources grow. Competitive dynamics might also select against consciousness, happiness, and human-valued traits (music, play, philosophy) entirely — "evolution" doesn't imply "progress."

Key Terms¶

Malthusian principle · superorganism · observation selection effect · precommitment (bargaining) · second transition

Case Summary¶

US horse population: 26 million (1915) → 2 million (early 1950s) once engines substituted for rather than complemented equine labor. Horses had no alternative employment to earn their keep — the direct analogy for human labor once machine intelligence is both cheaper and more capable across virtually all jobs.

Application Checklist¶

[ ] Distinguish "labor substitute" from "labor complement" when assessing automation's economic effect — substitutes crash wages, complements raise them
[ ] Don't assume competitive/evolutionary selection trends toward anything humans would recognize as progress or well-being
[ ] Check whether a trait (signaling, play, display) still serves its original function in a new environment before assuming it persists
[ ] Recognize that "many competing powers" is not automatically stable — a second transition or new coordination technology can reconsolidate into a singleton

Self-Test¶

Why might capital ownership and political mobilization only delay, not prevent, a Malthusian outcome for a growing population of digital minds?
Why does Bostrom think observing life's long history on Earth is weak evidence that intelligent life was "likely" to evolve?
How could an AI's ability to make an unbreakable, publicly verifiable precommitment change the outcome of a bargaining negotiation compared to human negotiators?