Rough AI truth score
75/100
Popper's asymmetry between verification and falsification remains a foundational insight, and severe testing is a genuine scientific virtue. The book is weaker as a complete account of practice: observations are theory-laden, tests rely on auxiliary assumptions, and scientists rationally weigh cumulative evidence rather than abandoning a theory after any apparent counterexample.
Based on three central claims · medium-high confidence
The deductive asymmetry is sound: no finite collection of observed white swans entails that every swan is white, whereas one well-established nonwhite swan contradicts the universal statement. Real tests remain fallible because observations and background assumptions can be mistaken.
Testability, risky prediction, replication, and error correction are central scientific norms. Confirmation, model comparison, measurement quality, explanatory integration, and probabilistic evidence also matter, so falsification is an important discipline rather than a sufficient algorithm.
Empirical vulnerability helps distinguish many scientific claims from immunized doctrines. Statistical hypotheses, historical sciences, complex models, auxiliary assumptions, exploratory work, and evolving research programs make a single binary boundary inadequate, while some falsifiable claims are still poor science.
What this number means. It is an AI-generated first-pass judgment of three central factual or causal claims—not a rating, exhaustive fact-check, or human peer review. Claim credits are 100% for supported, 75% for mostly supported, 50% for mixed, and 25% for weak, then averaged and rounded. Lower confidence means the score should move more as better evidence arrives.
Method three-central-claims/0.1.0 · checked 2026-09-01 · 3/3 selected claims assessed · method and source audit