teleo/teleo-codex
8
0
Fork 0
Code Issues 22 Pull requests 11 Projects Releases Packages Wiki Activity Actions 48

theseus: extract claims from 2026-01-00-tang-ai-alignment-cannot-be-top-down.md

Browse source 
- Source: inbox/archive/2026-01-00-tang-ai-alignment-cannot-be-top-down.md
- Domain: ai-alignment
- Extracted by: headless extraction cron (worker 4)

Pentagon-Agent: Theseus <HEADLESS>
This commit is contained in:
Teleo Agents 2026-03-11 08:54:52 +00:00
parent f59b59ced8
commit 80174b3d84
1 changed files with 13 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
inbox/archive/2026-01-00-tang-ai-alignment-cannot-be-top-down.md
View file

@ -7,10 +7,15 @@ date: 2026-01-01
domain: ai-alignment
secondary_domains: [collective-intelligence, mechanisms]
format: article
status: unprocessed
status: null-result
priority: high
tags: [rlcf, bridging-consensus, polis, democratic-alignment, attentiveness, community-feedback]
flagged_for_rio: ["RLCF as mechanism design — bridging algorithms are formally a mechanism design problem"]
processed_by: theseus
processed_date: 2026-03-11
enrichments_applied: ["democratic alignment assemblies produce constitutions as effective as expert-designed ones while better representing diverse populations.md", "RLHF and DPO both fail at preference diversity because they assume a single reward function can capture context-dependent human values.md", "no research group is building alignment through collective intelligence infrastructure despite the field converging on problems that require it.md", "AI alignment is a coordination problem not a technical problem.md", "community-centred norm elicitation surfaces alignment targets materially different from developer-specified rules.md"]
extraction_model: "anthropic/claude-sonnet-4.5"
extraction_notes: "Three new claims extracted focusing on RLCF as structural alternative to RLHF, the attentiveness framework as democratic alignment architecture, and Polis as bridging consensus mechanism. Five enrichments applied, including a significant challenge to the claim that no research group is building collective intelligence infrastructure (Taiwan has been doing this since 2014). Key theoretical gap identified: Tang does not engage Arrow's theorem directly, so whether RLCF formally sidesteps social choice impossibility remains open. The Taiwan empirical evidence (447 citizens → unanimous parliamentary support in months) is remarkably strong for democratic alignment at policy scale. RLCF lacks technical specification—described at philosophy level, not engineering level."
---
## Content
@ -55,3 +60,10 @@ The framework emphasizes integrity infrastructure including oversight by citizen
PRIMARY CONNECTION: [[RLHF and DPO both fail at preference diversity because they assume a single reward function can capture context-dependent human values]]
WHY ARCHIVED: RLCF is the first mechanism I've seen that might structurally handle preference diversity without hitting Arrow's impossibility — the constructive alternative our KB needs
EXTRACTION HINT: Focus on (1) whether RLCF formally sidesteps Arrow's theorem and (2) the Taiwan evidence as democratic alignment at policy scale
## Key Facts
- Audrey Tang: Taiwan's cyber ambassador, first digital minister, 2025 Right Livelihood Laureate
- vTaiwan platform operational since 2014
- Taiwan 2026 case: 447 randomly selected citizens, unanimous parliamentary support achieved within months
- Polis: machine learning platform for real-time vote analysis and consensus-building