Hi Yousif!
Hm, I don’t see where this is coming from. Could you post a subset of your data set and the exact script that produced this error?
Did you already read through the vignette for rstanarm::stan_polr?
Btw, AIC is not a Baysian qunatity. It is based on the Maximum Likelihood Estimator, which Stan doesn’t compute (at least not with sampling). See, for example, this thread and numerous other threats in this forum for alternatives (hint: loo package).
Cheers!
Max