# Teleox.ai - machine-readable summary

## Who

Teleox.ai is Chris Royse's site for leapable.ai (https://leapable.ai),
private AI memory, provenance receipts, and Teleological Constellation
Training. It points to the leapable.ai YouTube build archive, field notes,
paper, measurements, proof files, and a direct contact path.

## Current Standard

- Name: The Constellation Standard
- Version: v0.1
- Published: 2026-04-25
- Canonical URL: https://teleox.ai/standard/v0.1
- PDF: https://teleox.ai/standard/v0.1.pdf

## What the site is

- A direct offer: turn expert archives and private corpora into AI context,
  source trails, labels, eval targets, and checks.
- A field-notes blog that translates leapable.ai YouTube videos, PAPER.md, alldata.md,
  and system research notes into concise context for creators, companies, and
  AI lab readers.
- A public paper, Standard, measured rows, artifacts, citations, and date trail.
- A Frontier Evidence Map with five axes and no composite score.
- Receipt tools: Sign writes demo receipts; Verify reads demo receipt status.
- A centroid registry for domain-specific output checks.
- A governance page for conflicts, scope guards, and leapable.ai operations.

## What the site is not

- Not the full leapable.ai app UI. leapable.ai is the operational context product.
- Not a social surface.
- Not a runtime database on teleox.ai in v1.
- Not a native chain integration.
- Not a generic company brochure. Chris Royse is the builder; Teleox.ai is the
  focused product, proof, and contact surface.

## Scope guards

SG-Case3:
  "0.961 mean WavLM-SECS (max 0.975, N=10, 1 speaker, English,
  within-WavLM-family). ECAPA-TDNN EXP-1 cross-encoder validation
  pending. Not a parity claim against any published TTS system."

SG-91:
  "up to 91 supervisory signals per input, from n · (N + C(N,2)) at
  N=13, as a constructive upper bound. Pairwise-MI (EXP-2) audit
  pending. Realized floor will be published alongside the ceiling."

SG-Shumailov:
  "Structurally outside the recursion regime assumed by Shumailov et al.
  (2024). Scope argument — not a refutation of the collapse result.
  One constructive path for deriving labels from fixed real data."

SG-Gtau:
  "G_τ is a per-output cosine-threshold predicate over embedder outputs.
  This is a different property class than scalar-reward alignment (e.g.,
  RLHF), not a benchmark contest with it."

## Public routes

- https://teleox.ai/
- https://teleox.ai/blog
- https://teleox.ai/blog/meaning-compression-data-wall
- https://teleox.ai/blog/context-engineering-formula
- https://teleox.ai/blog/swappable-data-brains
- https://teleox.ai/blog/thirteen-embedder-context-graph
- https://teleox.ai/blog/document-provenance-pipeline
- https://teleox.ai/blog/zero-trust-agentic-computing
- https://teleox.ai/blog/agentic-engineering-workflow
- https://teleox.ai/blog/multi-agent-search-blueprint
- https://teleox.ai/blog/teleological-constellation-training
- https://teleox.ai/blog/derived-data-abundance-counting-identity
- https://teleox.ai/blog/third-path-around-the-data-wall
- https://teleox.ai/blog/runtime-constellation-guard
- https://teleox.ai/blog/clipcannon-video-understanding-dag
- https://teleox.ai/blog/ai-native-video-editor
- https://teleox.ai/blog/human-indistinguishable-clone-meeting-bot
- https://teleox.ai/blog/voice-cloning-0961-secs
- https://teleox.ai/blog/full-icl-voice-cloning
- https://teleox.ai/blog/shakespeare-lora-signal-density
- https://teleox.ai/blog/cross-lingual-period-style-transfer
- https://teleox.ai/blog/ocr-provenance-local-first-saas
- https://teleox.ai/blog/mcp-native-software-distribution
- https://teleox.ai/blog/air-gapped-document-intelligence
- https://teleox.ai/blog/zero-hallucination-legal-discovery
- https://teleox.ai/blog/perfect-citations-with-provenance
- https://teleox.ai/blog/ai-memory-without-hallucinations
- https://teleox.ai/blog/god-agent-local-file-workflow
- https://teleox.ai/blog/claude-local-superpowers
- https://teleox.ai/blog/forensic-driven-development
- https://teleox.ai/blog/game-theory-agents-for-claude-code
- https://teleox.ai/blog/multi-agent-paper-in-six-hours
- https://teleox.ai/blog/ai-proof-education-100-holes
- https://teleox.ai/blog/ai-team-communication-failure
- https://teleox.ai/blog/local-gpu-inference-economics
- https://teleox.ai/blog/ai-ethics-needs-tradeoffs
- https://teleox.ai/standard
- https://teleox.ai/standard/v0.1
- https://teleox.ai/standard/changelog
- https://teleox.ai/evidence-map
- https://teleox.ai/evidence-map/anthropic
- https://teleox.ai/evidence-map/openai
- https://teleox.ai/evidence-map/google-deepmind
- https://teleox.ai/evidence-map/meta
- https://teleox.ai/evidence-map/xai
- https://teleox.ai/sign
- https://teleox.ai/verify
- https://teleox.ai/feed
- https://teleox.ai/centroids
- https://teleox.ai/centroids/synthetic-media-impersonation
- https://teleox.ai/centroids/document-provenance
- https://teleox.ai/centroids/financial-risk-language
- https://teleox.ai/mark
- https://teleox.ai/mark/policy
- https://teleox.ai/method
- https://teleox.ai/measured
- https://teleox.ai/paper
- https://teleox.ai/infrastructure
- https://teleox.ai/artifacts
- https://teleox.ai/priority-date
- https://teleox.ai/pricing
- https://teleox.ai/governance
- https://teleox.ai/about
- https://teleox.ai/contact
- https://teleox.ai/citations

## Contact

Use https://teleox.ai/contact or email Chris directly with the corpus, target
task, useful proof, and sharing limits.

Email: chrisroyseai@gmail.com