TI Morse Field Scope Integration

Status

This page is website copy for a bounded Project Telos source-intake pass. The evidence label is SOURCE_RECEIPTS_MATCH: five requested videos have metadata and transcript receipts, and one Relentless channel list has a bounded flat-playlist receipt. The field lanes are INFERRED. Domain correctness remains UNVERIFIABLE_UNTIL_PRIMARY_SOURCE_OR_REPLAY.

Captured sources

Nuclear manufacturing

MATCH receipts. RZiM3Xfp-eY, "Manufacturing 1,000s of Nuclear Reactors | Isaiah Taylor, Valar Atomics".

Causal inference

MATCH receipts. 7lPWtFXsuzk, "Causal Inference with DAGs".

ARC-style AGI benchmarks

MATCH receipts. Vg6FBKTlfOw, "ARC-AGI-3 Explained by the Team That's Winning It".

Microscopy and materials

MATCH receipts. DyIQkqBXhS0, "Atomic Force Microscope high-speed video, selective stainless steel etching, bacteria, and more".

AI scale economics

MATCH receipts. bgWq678Oed4, "Why AI Companies Need Scale [Wading Through AI - Episode 7]".

Channel queue

MATCH receipt. First 12 flat-playlist entries from @ti_morse / Relentless.

Megatool lanes

Industrial Science Proof Packets

Gather source receipts, instrument/factory state, units, assumptions, safety boundaries, BuildLang/buildc simulation or measurement replay, and Crucible verdict.

Causal Research Workbench

Variables, DAG assumptions, adjustment sets, colliders, confounders, interventions, countergraphs, negative controls, synthetic-data replay, and claim cards.

Agentic Benchmark Foundry

Task source, observation budget, action budget, search policy, tool calls, hidden-test hygiene, verifier outcomes, score, and contamination controls.

Compute and Infrastructure Ledger

GPU, data-center, energy, cost, latency, utilization, model/eval config, artifact hashes, objective drift, and reproducible budget accounting.

First promotion target

The recommended first replayed demo is the Causal Research Workbench. It is small enough to make public, formal enough to test, and useful across medicine, biology, policy, economics, AI evaluation, and research review.

DAG source intake
-> typed DAG spec
-> adjustment-set check
-> countergraph / negative-control fixture
-> synthetic-data replay
-> claim card
-> Crucible verdict
-> Learning Forge lesson

Receipts

• Source ledger: demo/research/youtube-ti-morse-field-receipts.json.
• Receipt set seal: f7e6ebfa379e060d1a53bf3a3af1adb666aa6b9a3216897549810d7ea6139725.
• Crucible run: twenty-sixth-wave-ti-morse-field-run-2026-07-02.json.
• Crucible report: twenty-sixth-wave-ti-morse-field-report-2026-07-02.md.
• Learn prooflesson: twenty-sixth-wave-ti-morse-field-intake.prooflesson.json.
• Learn witness: 809ddaaa2826351ab1fc47f86b315d73131ffa8ab0e82c85402c8b1c9ac62324.

Do not infer

• Do not infer that the videos prove their domain claims.
• Do not infer that a channel queue proves market demand.
• Do not infer that transcript term counts prove technical content quality.
• Do not infer that BuildLang/buildc already implements the proposed causal, industrial, microscopy, or infrastructure receipts.
• Do not infer that the first promotion target is the only strategic target.