Skip to content

v1.1.0 — Legibility, companion site, and empirical spine

Choose a tag to compare

View all tags
@GunnarZarncke GunnarZarncke released this 08 Jul 10:09
v1.1.0
baf7d1c

Release Notes

Towards Superintelligence Alignment: Boundaries, Values, and Correction

v1.1.0 — 2026-07-08 — Legibility, companion site, and empirical spine

A consolidation release focused on external legibility (making the framework readable and checkable by outside researchers, funders, and policy readers), a full companion website, a new institutional-histories appendix, and four empirical experiment lines that stress-test bridge cruxes.

Reader-facing language and legibility

  • Cleaned up vocabulary Load-bearing jargon now leads with an operational paraphrase and keeps coined terms only as shorthand. Symbols, equations, and Lean predicate names are unchanged.
  • Operational glossary (Appendix E) rewritten and expanded. Each entry is paraphrase-first with a nearest field term delta and a concrete (often non-AI) example.
  • Front-matter orientation, README split, guided reading paths, and the About authorship note tightened for outside readers.

New appendix — Institutional histories (Appendix D / appM)

  • Mechanism-led historical case studies of institutional genesis, stabilization, entrenchment, and failure, with prior AI-governance literature credited; 24 sources in references/institutional-histories.bib; cross-refs in Appendix C and Chapters 27, 31, 34, 37.
  • Site overview hub with eleven case-study cards, a non-technical intro, and a full synced-text subroute; wired into the Funder/Policy and Philosopher guided paths.

Companion website (site/)

  • Astro publication layer: guided paths, concept and standalone-claim cards, chapter pages, a Lean-spine web layer with graph navigation, badges, an experiments page (YAML-synced), and in-browser PDF.

Empirical spine (experiments/)

  • Four in-repo simulation lines (toy, embedded, goal-agent, lab) that stress-test bridge cruxes with tentative, partial support only.
  • Lab-simulation line added and taken through multiple phases (graded red-team access tiers, unattributed-drift detection, blind detector generation, LLM red-team adapter, test batteries, adversarial-lineage gaming); goal-agent and embedded-sim lines extended; Lean spine gained Shannon entropy / mutual-information bounds and a successor-forgeability (MB10) formalization.

Housekeeping

  • src/ renamed to demos/;
  • bibliography enriched with canonical DOI/URL fields.
  • make check (structure, citations, bibliography summaries) passes.
Assets 3
Loading