The Jeffreys–Lindley (JL) paradox (also known as the Lindley paradox) reveals a foundational tension in statistical inference: When testing a point-null hypothesis H 0 : θ = θ 0 against a composite alternative in a continuous model, Bayesian and frequentist methods can yield starkly different conclusions as sample size increases. Specifically, frequentist p-values may strongly reject H 0 for large n even for negligible effects, while Bayesian posterior probabilities or Bayes factors may favor H 0 if the prior assigns nonzero mass to θ 0. This divergence is especially prominent in the era of large datasets, where such paradoxical results frequently arise and can undermine confidence in conventional statistical hypothesis testing.