The Half-Lives of Generative-AI Evidence: A 40-Record Audit, a Claim-Currency Framework, and a Reflexive Case of Frontier-Model-Assisted Research
2026-07-27 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors studied how old the AI models were at the time of research publication about generative AI systems, finding that the newest models were often several months old. They analyzed 40 research records from different sources and noted differences in model age depending on publication type. The paper also shows how the authors used an AI model themselves to assist quickly with research tasks, while still verifying all decisions carefully. They propose clearer ways to report AI model details and caution that model age and research claims are not the same. The title’s reference to “half-lives” is just a metaphor, not a scientific measurement.
Generative AIModel agePublication timingSupersessionPreprintsOpenAIResearch auditReflexive researchModel currencyHalf-life metaphor
Authors
Carlo Iacono
Abstract
Generative-AI evaluations can become historical before publication, yet calendar age does not affect every conclusion equally. This paper has two linked purposes. First, it audits a maximum-variation purposive corpus of 40 empirical records appearing between 18 July 2025 and 17 July 2026. The audit coded publication route, execution timing, model identity, age of the newest named generation or immutable snapshot, same-family supersession and refresh behaviour. At appearance, the newest named model was a median 281 days old (middle 50%: 75-478; range: 11-939). Median age was 395 days for 25 journal articles, 56 days for 14 preprints and 49 days for one laboratory report. Thirty-five records included a superseded family, seven supplied a precise dated identifier, three clearly refreshed model evidence, and one added a late sensitivity test. All 40 included an OpenAI system, a feature of this corpus rather than a prevalence estimate. The paper distinguishes model age from claim currency and proposes six reporting practices. Second, it treats its own two-day production process as a reflexive case of frontier-model-assisted research creation. GPT-5.6 Sol Pro in ChatGPT supported candidate discovery, source reconciliation, calculations, drafting and critique; the author checked sources, made all substantive decisions and accepts responsibility. This is a proof-of-practice, not a controlled estimate of productivity or quality. By applying its own Model Facts and model-currency statement, the paper shows how rapid AI-assisted research can be made inspectable without treating model output as independent validation. The title uses half-lives metaphorically; no universal decay rate is estimated.