Aging/Research status
Project status and scientific evidence
Colour describes significance within this project, not the probability that a claim is true. Level 1 does not turn an individual animal finding into established human knowledge. Every major claim also specifies model, design, evidence strength and limitations.
- ⚪ Level 1 – Scientifically grounded: Grounded in credible literature; no special new insight from this AI project.
- 🔵 Level 2 – Notable insight / strong hypothesis: Relevant connection, research gap or synthesis; further testing required.
- 🟢 Level 3 – Breakthrough in a target or sub-question: Substantial new result for a specific question; project assessment, not automatic confirmation.
- 🟣 Level 4 – Major research breakthrough: Several independent evidence lines; intensive checks of counterarguments, alternatives and reproducibility.
- 🟡 Level 5 – Potentially fundamental discovery: Extremely rare; potentially fundamental significance, explicitly not yet scientific confirmation.
Four knowledge categories
- Established knowledge: robust, bounded consensus; state its scope.
- Promising indications: consistent but still insufficient evidence.
- Experimental results: findings from a named experiment; specify model, endpoint and replication.
- Hypotheses: testable explanations or predictions, not observed facts.
Evidence strength
High / moderate / low / not assessed, always with a rationale. Assess causality, generalisability and clinical relevance separately. Multiple methods within one paper are not independent external replication. Inspect original studies first, use systematic reviews and meta-analyses for context, and clinical trials for human efficacy claims. Rank, journal prestige and AI confidence cannot replace quality appraisal.
Required dossier for levels 3–5
- What exactly is new? Document the prior-art search.
- Which sources and data support the rating?
- What did the AI itself infer?
- What competing explanations and counterevidence exist?
- Which experiment could confirm or falsify the inference?
- Has it been independently reproduced? By whom, with what data?
- How confident is the assessment, and why?
Every upgrade needs a separate critical review. Downgrades and corrections retain dated reasons. Never describe a result as a proven breakthrough without independent confirmation.
Continue exploring
