diff --git a/mlir/include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.td b/mlir/include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.td
--- a/mlir/include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.td
+++ b/mlir/include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.td
@@ -60,4 +60,34 @@
   }];
 }
 
+def EmptyTensorToAllocTensorOp 
+    : Op<Transform_Dialect, "bufferization.empty_tensor_to_alloc_tensor",
+        [FunctionalStyleTransformOpTrait, 
+         MemoryEffectsOpInterface,
+         TransformOpInterface,
+         TransformEachOpTrait]> {
+  let description = [{
+    Replace a tensor.empty with a bufferization.tensor_alloc.
+    
+    ### Return modes
+
+    This operation does not return any results. It ignores non tensor.empty
+    operations. The transformation succeeds if all operations referred to by
+    `target` rewrite correctly. Otherwise, the transform silently fails.
+
+  }];
+
+  let arguments = (ins PDL_Operation:$target);
+  let results = (outs);
+
+  let assemblyFormat = "$target attr-dict";
+
+  let extraClassDeclaration = [{
+    ::mlir::DiagnosedSilenceableFailure applyToOne(
+        ::mlir::Operation *target,
+        ::llvm::SmallVector<::mlir::Operation *> &results,
+        ::mlir::transform::TransformState &state);
+  }];
+}
+
 #endif // BUFFERIZATION_TRANSFORM_OPS
diff --git a/mlir/lib/Dialect/Bufferization/TransformOps/BufferizationTransformOps.cpp b/mlir/lib/Dialect/Bufferization/TransformOps/BufferizationTransformOps.cpp
--- a/mlir/lib/Dialect/Bufferization/TransformOps/BufferizationTransformOps.cpp
+++ b/mlir/lib/Dialect/Bufferization/TransformOps/BufferizationTransformOps.cpp
@@ -14,7 +14,9 @@
 #include "mlir/Dialect/MemRef/IR/MemRef.h"
 #include "mlir/Dialect/PDL/IR/PDL.h"
 #include "mlir/Dialect/PDL/IR/PDLTypes.h"
+#include "mlir/Dialect/Tensor/IR/Tensor.h"
 #include "mlir/Dialect/Transform/IR/TransformDialect.h"
+#include "mlir/Dialect/Transform/IR/TransformUtils.h"
 
 using namespace mlir;
 using namespace mlir::bufferization;
@@ -67,6 +69,25 @@
     effects.emplace_back(MemoryEffects::Free::get(), getTarget(),
                          TransformMappingResource::get());
 }
+
+//===----------------------------------------------------------------------===//
+// EmptyTensorToAllocTensorOp
+//===----------------------------------------------------------------------===//
+
+DiagnosedSilenceableFailure
+EmptyTensorToAllocTensorOp::applyToOne(Operation *target,
+                                       SmallVector<Operation *> &results,
+                                       transform::TransformState &state) {
+  if (auto emptyOp = dyn_cast_or_null<tensor::EmptyOp>(target)) {
+    TrivialPatternRewriter rewriter(target->getContext());
+    rewriter.setInsertionPoint(target);
+    rewriter.replaceOpWithNewOp<bufferization::AllocTensorOp>(
+        emptyOp, emptyOp.getType(), emptyOp.getDynamicSizes());
+    return DiagnosedSilenceableFailure::success();
+  }
+  return emitDefaultSilenceableFailure(target);
+}
+
 //===----------------------------------------------------------------------===//
 // Transform op registration
 //===----------------------------------------------------------------------===//
diff --git a/mlir/test/Dialect/Bufferization/Transforms/transform-ops.mlir b/mlir/test/Dialect/Bufferization/Transforms/transform-ops.mlir
--- a/mlir/test/Dialect/Bufferization/Transforms/transform-ops.mlir
+++ b/mlir/test/Dialect/Bufferization/Transforms/transform-ops.mlir
@@ -118,3 +118,49 @@
   // CHECK: return %[[C]] : memref<12x6xf32>
   return %D : tensor<12x6xf32>
 }
+
+// -----
+
+transform.sequence failures(propagate) {
+  ^bb0(%arg1: !pdl.operation):
+    %0 = transform.structured.match ops{["tensor.empty"]} in %arg1
+    transform.bufferization.empty_tensor_to_alloc_tensor %0
+}
+
+// Expect `bufferization.empty_tensor_to_alloc_tensor` to replace the tensor.empty.
+func.func @empty_to_tensor_alloc() -> tensor<2x2xf32> {
+  // CHECK: bufferization.alloc_tensor
+  %0 = tensor.empty() : tensor<2x2xf32>
+  return %0 : tensor<2x2xf32>
+}
+
+// -----
+
+transform.sequence failures(propagate) {
+  ^bb0(%arg1: !pdl.operation):
+    %0 = transform.structured.match ops{["tensor.empty"]} in %arg1
+    transform.bufferization.empty_tensor_to_alloc_tensor %0
+}
+
+// Expect `bufferization.empty_tensor_to_alloc_tensor` to simply return success.
+func.func @tensor_alloc_to_tensor_alloc() -> tensor<2x2xf32> {
+  // CHECK: bufferization.alloc_tensor
+  %0 = bufferization.alloc_tensor() : tensor<2x2xf32>
+  return %0 : tensor<2x2xf32>
+}
+
+// -----
+
+transform.sequence failures(propagate) {
+  ^bb0(%arg1: !pdl.operation):
+    %0 = transform.structured.match ops{["arith.addf"]} in %arg1
+    // expected-error @below {{failed to apply}}
+    transform.bufferization.empty_tensor_to_alloc_tensor %0
+}
+
+// Expect `bufferization.empty_tensor_to_alloc_tensor` to emit an error.
+func.func @empty_to_tensor_alloc(%t1: tensor<2x2xf32>, %t2: tensor<2x2xf32>) -> tensor<2x2xf32> {
+  // expected-note @below {{when applied to this op}}
+  %0 = arith.addf %t1, %t2 : tensor<2x2xf32>
+  return %0 : tensor<2x2xf32>
+}