"See help(type(self)) for accurate signature" in `__init__`'s docstring is misleading

Feature or enhancement

Let's say I have a simple class:

class A:
    def __init__(self, a):
        self.a = a

help(A) will produce:

Help on class A in module __main__:

class A(builtins.object)
 |  A(a)
 |
 |  Methods defined here:
 |
 |  __init__(self, a)
 |      Initialize self.  See help(type(self)) for accurate signature.
 |
 |  ----------------------------------------------------------------------
 |  Data descriptors defined here:
 |
 |  __dict__
 |      dictionary for instance variables
 |
 |  __weakref__
 |      list of weak references to the object

Note that the signature is already accurate. But, now I see this message: See help(type(self)) for accurate signature.. Let's try to do something about it.

Let's follow the instruction:

class A:
    def __init__(self, a):
        help(type(self))
        self.a = a

It will produce:

Help on class A in module __main__:

class A(builtins.object)
 |  A(a)
 |
 |  Methods defined here:
 |
 |  __init__(self, a)
 |      Initialize self.  See help(type(self)) for accurate signature.
 |
 |  ----------------------------------------------------------------------
 |  Data descriptors defined here:
 |
 |  __dict__
 |      dictionary for instance variables
 |
 |  __weakref__
 |      list of weak references to the object

Exactly the same signature. With exactly the same message.

Now, let's try a different example, an empty class:

class B: ...

help(B.__init__) will say:

Help on wrapper_descriptor:

__init__(self, /, *args, **kwargs) unbound builtins.object method
    Initialize self.  See help(type(self)) for accurate signature.

And help(B) will say (notice that there would be no __init__ at all):

Help on class B in module __main__:

class B(builtins.object)
 |  Data descriptors defined here:
 |
 |  __dict__
 |      dictionary for instance variables
 |
 |  __weakref__
 |      list of weak references to the object

Is this message See help(type(self)) for accurate signature really useful? When do users benefit from seeing it?

Reproduce the reported output with help(A), help(A.init), and an empty class B. Trace where the built-in init docstring and pydoc class output are generated, then identify the relevant regression-test area. Done means the misleading guidance is corrected or removed consistently and the help output is covered by tests.