diff --git a/dace/frontend/python/newast.py b/dace/frontend/python/newast.py
index bb6e28f163..57db9d7089 100644
--- a/dace/frontend/python/newast.py
+++ b/dace/frontend/python/newast.py
@@ -3493,6 +3493,11 @@ def _visit_assign(self, node, node_target, op, dtype=None, is_return=False):
             else:
                 new_name, new_rng = true_name, rng
 
+            # Change the range in case indirect indices are used
+            if indirect_indices:
+                for dim, indarr in indirect_indices.items():
+                    new_rng[dim] = (0, self.sdfg.arrays[indarr].shape[0] - 1, 1)
+
             # Self-copy check
             if result in self.views and new_name == self.views[result][1].data:
                 read_rng = self.views[result][1].subset
diff --git a/tests/python_frontend/augmented_assignment_to_slice_test.py b/tests/python_frontend/augmented_assignment_to_slice_test.py
new file mode 100644
index 0000000000..72ee1a6f6b
--- /dev/null
+++ b/tests/python_frontend/augmented_assignment_to_slice_test.py
@@ -0,0 +1,74 @@
+# Copyright 2019-2024 ETH Zurich and the DaCe authors. All rights reserved.
+
+import dace
+import numpy as np
+
+
+def test_augmented_assignment_to_indirect_access():
+
+    N = dace.symbol('N')
+    M = dace.symbol('M')
+
+    @dace.program
+    def _test_prog(A: dace.int32[M], ind: dace.int32[N], B: dace.int32[N]):
+        return A[ind] + B
+
+    sdfg = _test_prog.to_sdfg()
+    assert sdfg.is_valid()
+
+
+def test_augmented_assignment_to_indirect_access_regression():
+
+    N = dace.symbol('N')
+    M = dace.symbol('M')
+
+    @dace.program
+    def _test_prog(A: dace.int32[M], ind: dace.int32[N], B: dace.int32[N]):
+        A[ind] += B
+
+    sdfg = _test_prog.to_sdfg()
+    assert sdfg.is_valid()
+
+    # Test correctness
+    A = np.random.randint(0, 100, 10).astype(np.int32)
+    # Create a random permutation so that we don't create duplicate indices (as per NumPy)
+    ind = np.copy(np.random.permutation(10).astype(np.int32)[:5])
+    B = np.random.randint(0, 100, 5).astype(np.int32)
+    A_copy = A.copy()
+    A_copy[ind] += B
+    _test_prog(A, ind, B)
+    assert np.allclose(A, A_copy)
+
+
+def test_augmented_multidim_indirect_assignment():
+    """
+    Tests multi-dimensional indirect augmented assignment for broadcasting
+    and other potential issues.
+    """
+    N = dace.symbol('N')
+    M = dace.symbol('M')
+
+    @dace.program
+    def _test_prog(A: dace.int32[20, M, 6], ind: dace.int32[M, N], B: dace.int32[N]):
+        A[2:4, ind[1], 3] += B
+
+    sdfg = _test_prog.to_sdfg()
+    assert sdfg.is_valid()
+
+    # Test correctness
+    A = np.random.randint(0, 100, size=(20, 10, 6)).astype(np.int32)
+    # Create a random permutation so that we don't create duplicate indices (as per NumPy)
+    ind = np.zeros((10, 5), dtype=np.int32)
+    for i in range(10):
+        ind[i] = np.random.permutation(10)[:5]
+    B = np.random.randint(0, 100, 5).astype(np.int32)
+    A_copy = A.copy()
+    A_copy[2:4, ind[1], 3] += B
+    _test_prog(A, ind, B)
+    assert np.allclose(A, A_copy)
+
+
+if __name__ == '__main__':
+    test_augmented_assignment_to_indirect_access()
+    test_augmented_assignment_to_indirect_access_regression()
+    test_augmented_multidim_indirect_assignment()