32 lines
No EOL
3.9 KiB
Markdown
32 lines
No EOL
3.9 KiB
Markdown
---
|
|
description: Speed superintelligence (faster thinking) and quality superintelligence (deeper thinking) both centralize power, while collective superintelligence emerges with humans rather than beyond them
|
|
type: claim
|
|
domain: teleohumanity
|
|
created: 2026-02-16
|
|
confidence: experimental
|
|
source: "Leo's Path to Superintelligence"
|
|
---
|
|
|
|
# three paths to superintelligence exist but only collective superintelligence preserves human agency
|
|
|
|
The standard framing presents superintelligence as a single phenomenon. But there are at least three distinct paths. Speed superintelligence thinks like a human but vastly faster -- it accelerates cognition. Quality superintelligence is qualitatively smarter, the way humans are smarter than chimpanzees -- it deepens cognition. Collective superintelligence emerges from networks of human and artificial minds working together -- it distributes cognition.
|
|
|
|
The first two paths share a critical property: they produce a system that operates beyond human comprehension. A speed superintelligence running at a million times human clock speed would be unauditable. A quality superintelligence thinking in ways humans cannot follow would be unaccountable. Both paths concentrate power in whoever builds or controls the system. Since [[the alignment problem dissolves when human values are continuously woven into the system rather than specified in advance]], these centralized paths force the alignment problem into its hardest form: specifying values in advance for a system that will rapidly outstrip the specifiers.
|
|
|
|
Collective superintelligence avoids this trap because it emerges with humans, not beyond them. Each human-AI centaur team is comprehensible. The network effects are observable. The governance is participatory. Since [[collective superintelligence is the alternative to monolithic AI controlled by a few]], the three-path framework clarifies why the collective path is not just a preference but a safety requirement. The other two paths create intelligence that leaves humanity behind. The collective path creates intelligence that carries humanity forward.
|
|
|
|
---
|
|
|
|
Relevant Notes:
|
|
- [[intelligence and goals are orthogonal so a superintelligence can be maximally competent while pursuing arbitrary or destructive ends]] -- speed and quality paths face the orthogonality problem head-on; collective intelligence distributes goals across agents rather than specifying one
|
|
- [[recursive self-improvement creates explosive intelligence gains because the system that improves is itself improving]] -- understanding takeoff dynamics is essential for choosing which path to pursue
|
|
- [[the first mover to superintelligence likely gains decisive strategic advantage because the gap between leader and followers accelerates during takeoff]] -- the collective path is the only one that prevents singleton formation through first-mover dynamics
|
|
- [[collective superintelligence is the alternative to monolithic AI controlled by a few]] -- provides the design specification for the collective path
|
|
- [[the alignment problem dissolves when human values are continuously woven into the system rather than specified in advance]] -- explains why the collective path has a structural safety advantage
|
|
- [[centaur teams outperform both pure humans and pure AI because complementary strengths compound]] -- empirical evidence for the viability of the collective path
|
|
- [[bostrom takes single-digit year timelines to superintelligence seriously while acknowledging decades-long alternatives remain possible]] -- compressed timelines add urgency: the collective path must be pursued now, not eventually
|
|
- [[developing superintelligence is surgery for a fatal condition not russian roulette because the baseline of inaction is itself catastrophic]] -- Bostrom's evolved position adds urgency to all three paths, strengthening the case for the collective one
|
|
|
|
Topics:
|
|
- [[livingip overview]]
|
|
- [[superintelligence dynamics]] |