diff --git a/runtime/core/portable_type/tensor_impl.cpp b/runtime/core/portable_type/tensor_impl.cpp
index ede5a3d4101..64ecb73a726 100644
--- a/runtime/core/portable_type/tensor_impl.cpp
+++ b/runtime/core/portable_type/tensor_impl.cpp
@@ -12,6 +12,7 @@
 #include <cstdint>
 
 #include <c10/util/irange.h>
+#include <c10/util/safe_numerics.h>
 
 #include <executorch/runtime/core/exec_aten/util/dim_order_util.h>
 #include <executorch/runtime/core/exec_aten/util/scalar_type_util.h>
@@ -38,7 +39,9 @@ ssize_t compute_numel(const TensorImpl::SizesType* sizes, ssize_t dim) {
         "Size must be non-negative, got %zd at dimension %zd",
         static_cast<ssize_t>(sizes[i]),
         i);
-    numel *= sizes[i];
+    bool overflow =
+        c10::mul_overflows(numel, static_cast<ssize_t>(sizes[i]), &numel);
+    ET_CHECK_MSG(!overflow, "numel overflowed");
   }
   return numel;
 }
@@ -66,7 +69,11 @@ TensorImpl::TensorImpl(
 }
 
 size_t TensorImpl::nbytes() const {
-  return numel_ * elementSize(type_);
+  size_t result;
+  bool overflow = c10::mul_overflows(
+      static_cast<size_t>(numel_), elementSize(type_), &result);
+  ET_CHECK_MSG(!overflow, "nbytes overflowed");
+  return result;
 }
 
 // Return the size of one element of the tensor
diff --git a/runtime/core/portable_type/test/tensor_impl_test.cpp b/runtime/core/portable_type/test/tensor_impl_test.cpp
index 0b8ae05f4da..08b26c64e09 100644
--- a/runtime/core/portable_type/test/tensor_impl_test.cpp
+++ b/runtime/core/portable_type/test/tensor_impl_test.cpp
@@ -10,6 +10,7 @@
 #include <executorch/runtime/core/portable_type/tensor_impl.h>
 
 #include <gtest/gtest.h>
+#include <cstdint>
 #include <random>
 
 #include <executorch/runtime/core/exec_aten/util/tensor_util.h>
@@ -449,3 +450,11 @@ TEST_F(TensorImplTest, TestResizingTensorToZeroAndBack) {
   EXPECT_GT(t.numel(), 0);
   EXPECT_EQ(t.data(), data);
 }
+
+TEST_F(TensorImplTest, TestNbytesOverflow) {
+  SizesType sizes[3] = {
+      static_cast<SizesType>(1 << 21),
+      static_cast<SizesType>(1 << 21),
+      static_cast<SizesType>(1 << 21)};
+  ET_EXPECT_DEATH(TensorImpl t(ScalarType::Float, 3, sizes), "");
+}