diff --git a/papers/2024/GPAM/README.md b/papers/2024/GPAM/README.md
index 91fd73bb..b91c00b2 100644
--- a/papers/2024/GPAM/README.md
+++ b/papers/2024/GPAM/README.md
@@ -103,8 +103,8 @@ loaded_model = tf.keras.models.load_model(
 
 How to use with [sedpack](https://github.com/google/sedpack) coming soon (if
 you already downloaded the datasets four small json files per dataset will be
-needed, if you used `gsutil rsync` to download, these will be downloaded very
-fast).
+needed, if you used `gcloud storage rsync` to download, these will be downloaded
+very fast).
 
 For now see the tutorial
 [https://google.github.io/sedpack/tutorials/sca/gpam/](https://google.github.io/sedpack/tutorials/sca/gpam/).
diff --git a/papers/datasets/ECC/GPAM/README.md b/papers/datasets/ECC/GPAM/README.md
index 28b9671a..3e14528e 100644
--- a/papers/datasets/ECC/GPAM/README.md
+++ b/papers/datasets/ECC/GPAM/README.md
@@ -14,13 +14,13 @@ Learning](https://github.com/google/scaaml/tree/main/papers/2024/GPAM).
 
 ```bash
 # List available datasets
-gsutil ls gs://scaaml-public/datasets/ECC/GPAM
+gcloud storage ls gs://scaaml-public/datasets/ECC/GPAM
 # Show size of the directory
-gsutil du -sh gs://scaaml-public/datasets/ECC/GPAM
+gcloud storage du --summarize --readable-sizes gs://scaaml-public/datasets/ECC/GPAM
 
 # Download all (beware that the multiprocessing option -m can saturate your
 # network connection). Or download just the dataset you want.
-gsutil -m rsync -r gs://scaaml-public/datasets/ECC/GPAM .
+gcloud storage rsync --recursive gs://scaaml-public/datasets/ECC/GPAM .
 ```
 
 ## How to load and use a dataset
diff --git a/project-words.txt b/project-words.txt
index 997c3ca2..3cc597e5 100644
--- a/project-words.txt
+++ b/project-words.txt
@@ -109,7 +109,7 @@ explainability
 figsize
 firstrow
 frombuffer
-gsutil
+gcloud storage
 hardcopyarea
 howpublished
 htmlcov