diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0f47c903d54..e451cc74fca 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -99,6 +99,7 @@ This release is compatible with NumPy 2.5.
 * Fixed `dpnp.cumsum`, `dpnp.cumprod`, and their `nan`/`cumulative_*` variants (including `dpnp.tensor.cumulative_sum`/`cumulative_prod`) silently returning incorrect results when accumulating along an axis of an array with more than one row [#3063](https://github.com/IntelPython/dpnp/pull/3063)
 * Fixed `dpnp.einsum` returning a result whose memory layout differs from NumPy for the default `order="K"`, and ignoring `out` and `order` for a contraction over a size-0 dimension [#3058](https://github.com/IntelPython/dpnp/pull/3058)
 * Fixed operations on a boolean array whose bytes are not `0x00`/`0x01` [#3055](https://github.com/IntelPython/dpnp/pull/3055)
+* Fixed the `dpnp.ndarray` constructor returning a view at the wrong address [#3068](https://github.com/IntelPython/dpnp/pull/3068)
 
 ### Security
 
diff --git a/dpnp/dpnp_array.py b/dpnp/dpnp_array.py
index b225fb2c732..3d0b3631d01 100644
--- a/dpnp/dpnp_array.py
+++ b/dpnp/dpnp_array.py
@@ -127,9 +127,26 @@ def __init__(
             # or as USM memory allocation
             if isinstance(buffer, dpnp_array):
                 buffer = buffer.get_array()
-                offset += buffer._element_offset
 
-            if dtype is None and hasattr(buffer, "dtype"):
+            if isinstance(buffer, dpt.usm_ndarray):
+                if dtype is None:
+                    dtype = buffer.dtype
+
+                # `buffer._element_offset` is expressed in units of the
+                # buffer's own dtype, while `offset` is interpreted in units
+                # of `dtype`, so the displacement has to be rescaled through
+                # bytes whenever the two itemsizes differ
+                byte_offset = buffer._element_offset * buffer.itemsize
+                new_itemsize = dpnp.dtype(dtype).itemsize
+                add_offset, rem = divmod(byte_offset, new_itemsize)
+                if rem != 0:
+                    raise ValueError(
+                        "The offset of the buffer's data in memory is not "
+                        "a multiple of the requested dtype size and so the "
+                        "requested view is not possible"
+                    )
+                offset += add_offset
+            elif dtype is None and hasattr(buffer, "dtype"):
                 dtype = buffer.dtype
         else:
             buffer = usm_type
diff --git a/dpnp/tests/test_ndarray.py b/dpnp/tests/test_ndarray.py
index f30317605aa..58916545823 100644
--- a/dpnp/tests/test_ndarray.py
+++ b/dpnp/tests/test_ndarray.py
@@ -529,6 +529,68 @@ def test_nonzero_offset_buffer_ctor(self):
         assert_array_equal(ia.view(), expected)
         assert_array_equal(ia.view(dpnp.uint32), expected.view(numpy.uint32))
 
+    @pytest.mark.parametrize(
+        "src_dt, new_dt",
+        [
+            (dpnp.complex64, dpnp.uint16),
+            (dpnp.complex128, dpnp.float64),
+            (dpnp.float64, dpnp.float32),
+            (dpnp.int64, dpnp.int8),
+            (dpnp.int32, dpnp.int16),
+            (dpnp.int16, dpnp.int64),
+        ],
+    )
+    def test_nonzero_offset_buffer_ctor_dtype_mismatch(self, src_dt, new_dt):
+        # the element offset of the `buffer=` array is expressed in units of
+        # the buffer's own dtype and has to be rescaled when the requested
+        # dtype has a different itemsize
+        base = dpnp.arange(32, dtype=src_dt)
+        sl = base[8:]
+
+        byte_offset = 8 * dpnp.dtype(src_dt).itemsize
+        size = (base.nbytes - byte_offset) // dpnp.dtype(new_dt).itemsize
+
+        ia = dpnp.ndarray((size,), dtype=new_dt, buffer=sl)
+        assert ia.data.ptr == sl.data.ptr
+        assert_array_equal(ia, dpnp.asnumpy(sl).view(new_dt))
+
+    def test_nonzero_offset_buffer_ctor_usm_ndarray(self):
+        # the same rescaling applies when `buffer=` is a bare usm_ndarray
+        # rather than a dpnp.ndarray
+        base = dpnp.arange(16, dtype=dpnp.complex64)
+        sl = base[4:]
+        usm_sl = sl.get_array()
+
+        for dt in [dpnp.complex64, dpnp.uint16, dpnp.float32]:
+            size = usm_sl.nbytes // dpnp.dtype(dt).itemsize
+            ia = dpnp.ndarray((size,), dtype=dt, buffer=usm_sl)
+            assert ia.data.ptr == sl.data.ptr
+
+        # and the dtype still defaults to the buffer's one
+        ia = dpnp.ndarray((12,), buffer=usm_sl)
+        assert ia.dtype == base.dtype
+        assert ia.data.ptr == sl.data.ptr
+
+    def test_nonzero_offset_buffer_ctor_write_through(self):
+        # a write through the dtype-mismatched view must land in the parent
+        # allocation at the offset the buffer points at
+        base = dpnp.zeros(16, dtype=dpnp.complex64)
+        sl = base[8:]
+
+        ia = dpnp.ndarray((16,), dtype=dpnp.float32, buffer=sl)
+        ia[:] = 1
+
+        expected = numpy.zeros(16, dtype=numpy.complex64)
+        expected[8:] = 1 + 1j
+        assert_array_equal(base, expected)
+
+    def test_misaligned_offset_buffer_ctor_error(self):
+        base = dpnp.arange(16, dtype=dpnp.int16)
+        # the buffer starts at a byte offset of 6, which is not addressable
+        # with an itemsize of 8
+        with pytest.raises(ValueError, match="not a multiple"):
+            dpnp.ndarray((3,), dtype=dpnp.int64, buffer=base[3:])
+
     def test_misaligned_offset_error(self):
         ia = dpnp.arange(10, dtype=dpnp.int16)
         # numpy supports such a view, but usm_ndarray cannot address memory