Rigor
Limits and open points
A good scientific showcase should not only display successes. It must say what is closed, what is conditional, what is numerically validated, and what still awaits external verification.
Physical bridges
BA bridges remain structuring commitments. Even when strongly motivated, they must stay distinct from algebraic identities.
Observational QG
The geometric chain is advanced, but fine observational identification of Kerr/QNM modes remains a testing edge.
Chemical corrections
IE/EA are promising, but every canonical correction must remain derived by continuous PT functions, not free coefficients.
Cosmology
The dark-sector reading must clearly separate derivation, interpretation, and observational prediction.
External review
The site presents a scientific program under construction. It still needs independent confrontation, critical review, and external reproduction.
Use of AI
The Theory of Persistence is first and foremost the product of human thought: its intuitions, principles, theoretical commitments, and overall direction were carried by its author. Large language models, especially Claude and ChatGPT, nevertheless played a substantial tool role in the development of the program.
They helped accelerate the writing of calculation and verification scripts, suggested mathematical tools the author did not always know, explained how to use them, and helped structure or draft some texts, pages, and articles. This use is explicit: it is part of contemporary research tooling, alongside computing environments, scientific libraries, and writing assistants.
This does not mean that these models understand the theory as a whole. Trained on established frameworks, including the Standard Model and conventional formalisms, they can help locally by checking, coding, reformulating, or suggesting analogies; but global coherence, interpretation of results, and responsibility for claims remain human.
A verifiable ambition
PT makes a strong claim: explaining a great deal with very little. That claim only matters if it remains testable. Its strength comes from the rigidity of its constraints, the convergence of its results, and its willingness to show where experiment or computation can still put it under pressure.