Describe the issue:

When you create an array with dtype=np.bool, mypy says the returned array's dtype is np.dtype[np.bool[Literal[False]]] despite the fact that it's an array, not a literal, and may contain values other than False. Pyright says the returned array's dtype is np.dtype[np.bool[bool]], which makes more sense to me because the array items are bools but they are not necessarily False.

I also filed a bug about TypeError when subscripting np.bool at run time: #29247

Reproduce the code example:

import numpy as np
x = np.array([True, False], np.bool)
reveal_type(x)

Error message:

# mypy 1.16.1
test.py:3: note: Revealed type is "numpy.ndarray[builtins.tuple[Any, ...], numpy.dtype[numpy.bool[Literal[False]]]]"
Success: no issues found in 1 source file

# pyright 1.1.402
test.py:3:13 - information: Type of "x" is "ndarray[tuple[Any, ...], dtype[numpy.bool[builtins.bool]]]"
0 errors, 0 warnings, 1 information

Python and NumPy Versions:

numpy 2.3.0
python 3.11.9 (tags/v3.11.9:de54cf5, Apr 2 2024, 10:12:12) [MSC v.1938 64 bit (AMD64)]

Type-checker version and settings:

mypy 1.16.1, mypy test.py, no configuration options
pyright 1.1.402, pyright test.py, no configuration options

Additional typing packages.

No response