Jensen's alpha

Why can’t we use Jensen’s alpha for measuring the over/under-performance of un-diversified portfolios? I can’t seem to find a good explanation for this anywhere

Jensen’s alpha is derived from CAPM which assumes that all expected returns can be explained by systematic risk (market beta). Therefore, Jensen’s alpha might give a misleading measure of an un-diversified portfolio as it does not reward unsystematic risk.

yes but according to the theory, you get no returns for assuming unsystematic risk so I don’t see how this is an issue.