GLOSSARY
Glossary: Human Edge, Phoenix Soulfire™ and AI authorship
The working vocabulary of the site, defined in the words already used on it.
The thesis
- Human Edge
- The BJ Beyond thesis: AI can generate and data can measure, but human judgment decides what matters. The point is not to compete with AI at what it does best; the point is to keep the decision human where judgment still carries weight.
- Source: /human-edge/ →
- Phoenix Experiment
- A recovered record and a question: what can we verify when AI explains its own response? The archive can document the sequence. It cannot tell us why the model chose to generate the image.
- Source: /phoenix-experiment/ →
The method
- Phoenix Soulfire™
- The five-test human judgment layer created by BJ Beyond to check whether a text, a work or a plan holds up beyond what AI can generate. It is a judgment layer, not an art scorecard and not a numeric rating; it does not authenticate an artwork or establish a market valuation.
- Source: /method/ →
- Legacy
- What remains after the feed moves. Fail signal: built to be posted, not to last.
- Source: /method/ →
- Phoenix Simulator
- An illustrative BJ Beyond Labs simulator exploring possible distribution signals on X. It is not X’s actual algorithm, and it is not the Phoenix Soulfire™ method.
- Source: /phoenix/ →
AI and authorship
- Claudefishing
- A term introduced by Substack co-founder Chris Best for a mismatch between what a reader believes they are reading and how the text was actually made. In BJ Beyond’s reading, AI authorship is a production fact; Claudefishing is a disclosure failure.
- Source: /writing/claude-opus-5-vs-pangram/ →
- Detection and disclosure
- Detectors read the artifact. Disclosure explains the process. Neither can do the other’s job: a detector can flag a text that deserves closer attention, while a process statement can say who supplied the ideas, who wrote the prose, what the machine changed and who accepts responsibility.
- Source: /writing/what-an-ai-detector-can-prove-and-what-it-cant/ →