Pandas BUG: inconsistent creation of IntegerArray from boolean values

When creating a nullable integer array from boolean values, we are being inconsistent between a boolean numpy array and a object array with bools:

>>> pd.array(np.array([True, False]), dtype="Int64")  
<IntegerArray>
[1, 0]
Length: 2, dtype: Int64

>>> pd.array(np.array([True, False], dtype=object), dtype="Int64")  
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-33-bc23b09ea959> in <module>
----> 1 pd.array(np.array([True, False], dtype=object), dtype="Int64")

~/scipy/pandas/pandas/core/construction.py in array(data, dtype, copy)
    282     if is_extension_array_dtype(dtype):
    283         cls = cast(ExtensionDtype, dtype).construct_array_type()
--> 284         return cls._from_sequence(data, dtype=dtype, copy=copy)
    285 
    286     if dtype is None:

~/scipy/pandas/pandas/core/arrays/integer.py in _from_sequence(cls, scalars, dtype, copy)
    346     @classmethod
    347     def _from_sequence(cls, scalars, dtype=None, copy=False):
--> 348         return integer_array(scalars, dtype=dtype, copy=copy)
    349 
    350     @classmethod

~/scipy/pandas/pandas/core/arrays/integer.py in integer_array(values, dtype, copy)
    127     TypeError if incompatible types
    128     """
--> 129     values, mask = coerce_to_array(values, dtype=dtype, copy=copy)
    130     return IntegerArray(values, mask)
    131 

~/scipy/pandas/pandas/core/arrays/integer.py in coerce_to_array(values, dtype, mask, copy)
    210             "mixed-integer-float",
    211         ]:
--> 212             raise TypeError(f"{values.dtype} cannot be converted to an IntegerDtype")
    213 
    214     elif is_bool_dtype(values) and is_integer_dtype(dtype):

TypeError: object cannot be converted to an IntegerDtype