diff --git a/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp b/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
index 3c07c1d..00b06f6 100644
--- a/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMMatrixMultiplyNativeKernel.cpp
@@ -65,6 +65,8 @@
     ARM_COMPUTE_RETURN_ERROR_ON(rhs_info.k0 > 16);
     ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.m0 < 1 || lhs_info.m0 > 8);
     ARM_COMPUTE_RETURN_ERROR_ON_MSG(((rhs_info.n0 & (rhs_info.n0 - 1)) && rhs_info.n0 != 3), "Only 2,3,4,8,16 are supported for n0");
+    ARM_COMPUTE_RETURN_ERROR_ON_MSG((gemm_info.reinterpret_input_as_3d || gemm_info.depth_output_gemm3d != 0) && (input2 != nullptr)
+                                    && (!gemm_info.broadcast_bias), "Bias addition only supported with broadcast mode in case the input or output has to be reinterpreted as 3D");
 
     const unsigned int m = gemm_info.m;
     const unsigned int n = gemm_info.n;