diff --git a/docs/guide/container_component.md b/docs/guide/container_component.md
index 67449cc7b9..04fcab136e 100644
--- a/docs/guide/container_component.md
+++ b/docs/guide/container_component.md
@@ -70,8 +70,8 @@ grep_component = tfx.dsl.components.create_container_component(
     parameters={
         'pattern': str,
     },
-    # The component code uses gsutil to upload the data to Google Cloud Storage, so the
-    # container image needs to have gsutil installed and configured.
+    # The component code uses gcloud storage to upload the data to Google Cloud Storage, so the
+    # container image needs to have gcloud storage installed and configured.
     image='google/cloud-sdk:278.0.0',
     command=[
         'sh', '-exc',
@@ -83,13 +83,13 @@ grep_component = tfx.dsl.components.create_container_component(
           filtered_text_path=$(mktemp)
 
           # Getting data into the container
-          gsutil cp "$text_uri" "$text_path"
+          gcloud storage cp "$text_uri" "$text_path"
 
           # Running the main code
           grep "$pattern" "$text_path" >"$filtered_text_path"
 
           # Getting data out of the container
-          gsutil cp "$filtered_text_path" "$filtered_text_uri"
+          gcloud storage cp "$filtered_text_path" "$filtered_text_uri"
         ''',
         '--pattern', tfx.dsl.placeholders.InputValuePlaceholder('pattern'),
         '--text', tfx.dsl.placeholders.InputUriPlaceholder('text'),
diff --git a/docs/tutorials/tfx/cloud-ai-platform-pipelines.md b/docs/tutorials/tfx/cloud-ai-platform-pipelines.md
index 40977a0d05..94979aab69 100644
--- a/docs/tutorials/tfx/cloud-ai-platform-pipelines.md
+++ b/docs/tutorials/tfx/cloud-ai-platform-pipelines.md
@@ -359,7 +359,7 @@ The notebook will upload our sample data to GCS bucket so that we can use it in
 our pipeline later.
 
 ```python
-!gsutil cp data/data.csv gs://{GOOGLE_CLOUD_PROJECT}-kubeflowpipelines-default/tfx-template/data/taxi/data.csv
+!gcloud storage cp data/data.csv gs://{GOOGLE_CLOUD_PROJECT}-kubeflowpipelines-default/tfx-template/data/taxi/data.csv
 ```
 
 The notebook then uses the `tfx pipeline create` command to create the pipeline.
diff --git a/docs/tutorials/tfx/gcp/vertex_pipelines_bq.ipynb b/docs/tutorials/tfx/gcp/vertex_pipelines_bq.ipynb
index bc35bdb777..0c4dbccc6a 100644
--- a/docs/tutorials/tfx/gcp/vertex_pipelines_bq.ipynb
+++ b/docs/tutorials/tfx/gcp/vertex_pipelines_bq.ipynb
@@ -637,7 +637,7 @@
       },
       "outputs": [],
       "source": [
-        "!gsutil cp {_trainer_module_file} {MODULE_ROOT}/"
+        "!gcloud storage cp {_trainer_module_file} {MODULE_ROOT}/"
       ]
     },
     {
diff --git a/docs/tutorials/tfx/gcp/vertex_pipelines_simple.ipynb b/docs/tutorials/tfx/gcp/vertex_pipelines_simple.ipynb
index 3c63483712..576bf8107f 100644
--- a/docs/tutorials/tfx/gcp/vertex_pipelines_simple.ipynb
+++ b/docs/tutorials/tfx/gcp/vertex_pipelines_simple.ipynb
@@ -387,7 +387,7 @@
       },
       "outputs": [],
       "source": [
-        "!gsutil cp gs://download.tensorflow.org/data/palmer_penguins/penguins_processed.csv {DATA_ROOT}/"
+        "!gcloud storage cp gs://download.tensorflow.org/data/palmer_penguins/penguins_processed.csv {DATA_ROOT}/"
       ]
     },
     {
@@ -407,7 +407,7 @@
       },
       "outputs": [],
       "source": [
-        "!gsutil cat {DATA_ROOT}/penguins_processed.csv | head"
+        "!gcloud storage cat {DATA_ROOT}/penguins_processed.csv | head"
       ]
     },
     {
@@ -607,7 +607,7 @@
       },
       "outputs": [],
       "source": [
-        "!gsutil cp {_trainer_module_file} {MODULE_ROOT}/"
+        "!gcloud storage cp {_trainer_module_file} {MODULE_ROOT}/"
       ]
     },
     {
diff --git a/docs/tutorials/tfx/stub_template.md b/docs/tutorials/tfx/stub_template.md
index d99fa455dd..59e5334dd0 100644
--- a/docs/tutorials/tfx/stub_template.md
+++ b/docs/tutorials/tfx/stub_template.md
@@ -143,7 +143,7 @@ tfx run create --pipeline-name $pipeline_name --endpoint=$endpoint \
 
 Use command `fg` to access the port-forwarding in the background then ctrl-C to
 terminate. You can delete the directory with recorded pipeline outputs using
-`gsutil -m rm -R $output_dir`.
+`gcloud storage rm --recursive $output_dir`.
 
 To clean up all Google Cloud resources used in this project, you can
 [delete the Google Cloud project](https://cloud.google.com/resource-manager/docs/creating-managing-projects#shutting_down_projects)
diff --git a/docs/tutorials/tfx/template.ipynb b/docs/tutorials/tfx/template.ipynb
index bf9592cbd4..16adb681ea 100644
--- a/docs/tutorials/tfx/template.ipynb
+++ b/docs/tutorials/tfx/template.ipynb
@@ -408,7 +408,7 @@
       },
       "outputs": [],
       "source": [
-        "!gsutil cp data/data.csv gs://{GOOGLE_CLOUD_PROJECT}-kubeflowpipelines-default/tfx-template/data/taxi/data.csv"
+        "!gcloud storage cp data/data.csv gs://{GOOGLE_CLOUD_PROJECT}-kubeflowpipelines-default/tfx-template/data/taxi/data.csv"
       ]
     },
     {
diff --git a/tfx/examples/chicago_taxi_pipeline/serving/start_model_server_aiplatform.sh b/tfx/examples/chicago_taxi_pipeline/serving/start_model_server_aiplatform.sh
index b8648b547f..b10ee68105 100644
--- a/tfx/examples/chicago_taxi_pipeline/serving/start_model_server_aiplatform.sh
+++ b/tfx/examples/chicago_taxi_pipeline/serving/start_model_server_aiplatform.sh
@@ -20,10 +20,10 @@ echo Running cloud serving...
 # Dir for model exported for serving, e.g., gs://<bucket>/serving_model/chicago_taxi_pipeline_kubeflow
 CLOUD_MODEL_DIR=$1
 
-gsutil ls $CLOUD_MODEL_DIR
+gcloud storage ls $CLOUD_MODEL_DIR
 
 # Pick out the directory containing the last trained model.
-MODEL_BINARIES=$(gsutil ls $CLOUD_MODEL_DIR \
+MODEL_BINARIES=$(gcloud storage ls $CLOUD_MODEL_DIR \
   | sort | grep '\/[0-9]*\/$' | tail -n1)
 
 echo latest model: $MODEL_BINARIES
diff --git a/tfx/examples/custom_components/container_components/download_grep_print_pipeline.py b/tfx/examples/custom_components/container_components/download_grep_print_pipeline.py
index e55cc1f20f..32f306876f 100644
--- a/tfx/examples/custom_components/container_components/download_grep_print_pipeline.py
+++ b/tfx/examples/custom_components/container_components/download_grep_print_pipeline.py
@@ -41,7 +41,7 @@
           wget "$0" -O "$output_data_path" || curl "$0" > "$output_data_path"
 
           # Getting data out of the container
-          gsutil cp "$output_data_path" "$output_data_uri"
+          gcloud storage cp "$output_data_path" "$output_data_uri"
         ''',
         placeholders.InputValuePlaceholder('url'),
         placeholders.OutputUriPlaceholder('data'),
@@ -74,13 +74,13 @@
           filtered_text_path=$(mktemp)
 
           # Getting data into the container
-          gsutil cp "$text_uri" "$text_path"
+          gcloud storage cp "$text_uri" "$text_path"
 
           # Running the main code
           grep "$pattern" "$text_path" >"$filtered_text_path"
 
           # Getting data out of the container
-          gsutil cp "$filtered_text_path" "$filtered_text_uri"
+          gcloud storage cp "$filtered_text_path" "$filtered_text_uri"
         ''',
         placeholders.InputValuePlaceholder('pattern'),
         placeholders.InputUriPlaceholder('text'),
@@ -105,7 +105,7 @@
           text_path=$(mktemp)
 
           # Getting data into the container
-          gsutil cp "$text_uri" "$text_path"
+          gcloud storage cp "$text_uri" "$text_path"
 
           # Running the main code
           cat "$text_path"
diff --git a/tfx/examples/custom_components/slack/README.md b/tfx/examples/custom_components/slack/README.md
index 7d5e621407..e3d6cd9d1b 100644
--- a/tfx/examples/custom_components/slack/README.md
+++ b/tfx/examples/custom_components/slack/README.md
@@ -114,7 +114,7 @@ pipeline in Google Cloud Platform for production.
 First prepare a gcs bucket for the pipeline run root:
 
 ```bash
-gsutil mb -p ${PROJECT_ID} gs://${BUCKET_NAME}
+gcloud storage buckets create gs://${BUCKET_NAME} --project=${PROJECT_ID}
 ```
 
 Let's copy the dataset CSV to the GCS where TFX ExampleGen will ingest it
diff --git a/tfx/examples/penguin/experimental/README.md b/tfx/examples/penguin/experimental/README.md
index 93b280e185..02402552ec 100644
--- a/tfx/examples/penguin/experimental/README.md
+++ b/tfx/examples/penguin/experimental/README.md
@@ -67,8 +67,8 @@ pipeline python file. Output can be found at `[BUCKET]/tfx`.
 
 <pre class="devsite-terminal devsite-click-to-copy">
 vi ~/penguin/experimental/penguin_pipeline_sklearn_gcp.py
-gsutil -m cp -r ~/penguin/data/* gs://[BUCKET]/penguin/data/
-gsutil -m cp ~/penguin/experimental/\*.py gs://[BUCKET]/penguin/experimental/
+gcloud storage cp --recursive ~/penguin/data/* gs://[BUCKET]/penguin/data/
+gcloud storage cp ~/penguin/experimental/\*.py gs://[BUCKET]/penguin/experimental/
 
 tfx pipeline create \
   --engine kubeflow \
@@ -77,7 +77,7 @@ tfx pipeline create \
 </pre>
 
 Note that
-`gsutil -m cp ~/penguin/experimental/*.py gs://[BUCKET]/penguin/experimental`
+`gcloud storage cp ~/penguin/experimental/*.py gs://[BUCKET]/penguin/experimental`
 will need to be run every time updates are made to the GCP example.
 Additionally, subsequent pipeline deployments should use `tfx pipeline update`
 instead of `tfx pipeline create`.
diff --git a/tfx/orchestration/kubeflow/test_utils.py b/tfx/orchestration/kubeflow/test_utils.py
index 71e81f24f3..b3f4dffde0 100644
--- a/tfx/orchestration/kubeflow/test_utils.py
+++ b/tfx/orchestration/kubeflow/test_utils.py
@@ -197,7 +197,7 @@ class HelloWorldComponent(BaseComponent):
       args=[
           'echo "hello ' +
           ph.exec_property('word') +
-          '" | gsutil cp - ' +
+          '" | gcloud storage cp - ' +
           ph.output('greeting')[0].uri
       ])
 
diff --git a/tfx/orchestration/kubeflow/v2/e2e_tests/artifact_value_placeholder_integration_test.py b/tfx/orchestration/kubeflow/v2/e2e_tests/artifact_value_placeholder_integration_test.py
index f5002c84f0..3e41497828 100644
--- a/tfx/orchestration/kubeflow/v2/e2e_tests/artifact_value_placeholder_integration_test.py
+++ b/tfx/orchestration/kubeflow/v2/e2e_tests/artifact_value_placeholder_integration_test.py
@@ -46,7 +46,7 @@ def _tasks_for_pipeline_with_artifact_value_passing():
             echo "Hello $message" >"$output_data_path"
 
             # Getting data out of the container
-            gsutil cp -r "$output_data_path" "$output_data_uri"
+            gcloud storage cp --recursive "$output_data_path" "$output_data_uri"
           """,
           placeholders.InputValuePlaceholder('message'),
           placeholders.OutputUriPlaceholder('data'),
diff --git a/tfx/orchestration/test_pipelines/download_grep_print_pipeline.py b/tfx/orchestration/test_pipelines/download_grep_print_pipeline.py
index 8af8aadc6f..aa440f2e56 100644
--- a/tfx/orchestration/test_pipelines/download_grep_print_pipeline.py
+++ b/tfx/orchestration/test_pipelines/download_grep_print_pipeline.py
@@ -41,7 +41,7 @@
           wget "$0" -O "$output_data_path" || curl "$0" > "$output_data_path"
 
           # Getting data out of the container
-          gsutil cp "$output_data_path" "$output_data_uri"
+          gcloud storage cp "$output_data_path" "$output_data_uri"
         ''',
         ph.exec_property('url'),
         ph.output('data')[0].uri,
@@ -74,13 +74,13 @@
           filtered_text_path=$(mktemp)
 
           # Getting data into the container
-          gsutil cp "$text_uri" "$text_path"
+          gcloud storage cp "$text_uri" "$text_path"
 
           # Running the main code
           grep "$pattern" "$text_path" >"$filtered_text_path"
 
           # Getting data out of the container
-          gsutil cp "$filtered_text_path" "$filtered_text_uri"
+          gcloud storage cp "$filtered_text_path" "$filtered_text_uri"
         ''',
         ph.exec_property('pattern'),
         ph.input('text')[0].uri,
@@ -105,7 +105,7 @@
           text_path=$(mktemp)
 
           # Getting data into the container
-          gsutil cp "$text_uri" "$text_path"
+          gcloud storage cp "$text_uri" "$text_path"
 
           # Running the main code
           cat "$text_path"