diff --git a/pandas/core/arrays/numpy_.py b/pandas/core/arrays/numpy_.py
index 6e4aa1a5efacf..52900d9b62dc2 100644
--- a/pandas/core/arrays/numpy_.py
+++ b/pandas/core/arrays/numpy_.py
@@ -395,6 +395,7 @@ def _cmp_method(self, other, op):
         if isinstance(other, PandasArray):
             other = other._ndarray
 
+        other = ops.maybe_prepare_scalar_for_op(other, (len(self),))
         pd_op = ops.get_array_op(op)
         other = ensure_wrapped_if_datetimelike(other)
         with np.errstate(all="ignore"):
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 7f970a72cb12c..8ac94111cca56 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -6804,6 +6804,7 @@ def _arith_method(self, other, op):
             return ops.frame_arith_method_with_reindex(self, other, op)
 
         axis = 1  # only relevant for Series other case
+        other = ops.maybe_prepare_scalar_for_op(other, (self.shape[axis],))
 
         self, other = ops.align_method_FRAME(self, other, axis, flex=True, level=None)
 
diff --git a/pandas/core/ops/__init__.py b/pandas/core/ops/__init__.py
index f6bde348888a1..9cccf1cff60a1 100644
--- a/pandas/core/ops/__init__.py
+++ b/pandas/core/ops/__init__.py
@@ -35,6 +35,7 @@
     comparison_op,
     get_array_op,
     logical_op,
+    maybe_prepare_scalar_for_op,
 )
 from pandas.core.ops.common import (  # noqa:F401
     get_op_result_name,
@@ -428,6 +429,7 @@ def f(self, other, axis=default_axis, level=None, fill_value=None):
 
         axis = self._get_axis_number(axis) if axis is not None else 1
 
+        other = maybe_prepare_scalar_for_op(other, self.shape)
         self, other = align_method_FRAME(self, other, axis, flex=True, level=level)
 
         if isinstance(other, ABCDataFrame):
diff --git a/pandas/core/ops/array_ops.py b/pandas/core/ops/array_ops.py
index 2ff93b203a001..39c6fa13f79a4 100644
--- a/pandas/core/ops/array_ops.py
+++ b/pandas/core/ops/array_ops.py
@@ -202,12 +202,11 @@ def arithmetic_op(left: ArrayLike, right: Any, op):
     ndarray or ExtensionArray
         Or a 2-tuple of these in the case of divmod or rdivmod.
     """
-
     # NB: We assume that extract_array and ensure_wrapped_if_datetimelike
-    #  has already been called on `left` and `right`.
+    #  have already been called on `left` and `right`,
+    #  and `maybe_prepare_scalar_for_op` has already been called on `right`
     # We need to special-case datetime64/timedelta64 dtypes (e.g. because numpy
     # casts integer dtypes to timedelta64 when operating with timedelta64 - GH#22390)
-    right = _maybe_upcast_for_op(right, left.shape)
 
     if (
         should_extension_dispatch(left, right)
@@ -439,7 +438,7 @@ def get_array_op(op):
         raise NotImplementedError(op_name)
 
 
-def _maybe_upcast_for_op(obj, shape: Shape):
+def maybe_prepare_scalar_for_op(obj, shape: Shape):
     """
     Cast non-pandas objects to pandas types to unify behavior of arithmetic
     and comparison operations.
diff --git a/pandas/core/series.py b/pandas/core/series.py
index 0715d892d7ed3..da25ca8e3f653 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -5315,6 +5315,7 @@ def _arith_method(self, other, op):
 
         lvalues = self._values
         rvalues = extract_array(other, extract_numpy=True, extract_range=True)
+        rvalues = ops.maybe_prepare_scalar_for_op(rvalues, lvalues.shape)
         rvalues = ensure_wrapped_if_datetimelike(rvalues)
 
         with np.errstate(all="ignore"):