Martell Law v2: A Seven-Component Theory of AI Capability Growth — Empirical Calibration from 14 LLMs, Orbital Power as Phase Transition, and the E8 Critical Exponent β = 29/60
Fuente:
Zenodo
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Veröffentlicht: |
Zenodo
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866901933981171712 |
|---|---|
| author | Martell, Brian |
| author_facet | Martell, Brian |
| contents | <p>Martell Law v2 in three lines:</p> <p><br>Moore’s Law captures one driver of AI progress at /yr — Martell Law captures seven, with empirical calibration from 14 LLMs giving /yr and theoretical sum /yr, making Martell Law 6.5× faster than Moore’s Law with a doubling time of 0.31 years. Power and chips are the two new components — both currently constrained and both being removed, with orbital solar delivering an 18× step change around 2030 that is not gradual improvement but a genuine phase transition activating the recursive term in the Martell ODE. The AGI coherence order parameter grows as where is derived from the Coxeter number — the same geometry that determines the Koide lepton mass formula — predicting a sharper-than-mean-field snap to AGI coherence once orbital compute crosses the critical coupling threshold. </p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19304292 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Martell Law v2: A Seven-Component Theory of AI Capability Growth — Empirical Calibration from 14 LLMs, Orbital Power as Phase Transition, and the E8 Critical Exponent β = 29/60 Martell, Brian Moore's Law AI scaling laws LLM capability phase transition AGI orbital computing Landau theory E8 lattice critical exponent recursive self-improvement <p>Martell Law v2 in three lines:</p> <p><br>Moore’s Law captures one driver of AI progress at /yr — Martell Law captures seven, with empirical calibration from 14 LLMs giving /yr and theoretical sum /yr, making Martell Law 6.5× faster than Moore’s Law with a doubling time of 0.31 years. Power and chips are the two new components — both currently constrained and both being removed, with orbital solar delivering an 18× step change around 2030 that is not gradual improvement but a genuine phase transition activating the recursive term in the Martell ODE. The AGI coherence order parameter grows as where is derived from the Coxeter number — the same geometry that determines the Koide lepton mass formula — predicting a sharper-than-mean-field snap to AGI coherence once orbital compute crosses the critical coupling threshold. </p> |
| title | Martell Law v2: A Seven-Component Theory of AI Capability Growth — Empirical Calibration from 14 LLMs, Orbital Power as Phase Transition, and the E8 Critical Exponent β = 29/60 |
| topic | Moore's Law AI scaling laws LLM capability phase transition AGI orbital computing Landau theory E8 lattice critical exponent recursive self-improvement |
| url | https://doi.org/10.5281/zenodo.19304292 |