diff --git a/detector/README.md b/detector/README.md
index 4b5c225..0c41572 100644
--- a/detector/README.md
+++ b/detector/README.md
@@ -12,13 +12,13 @@ For motivations and discussions regarding the release of this detector model, pl
 Download the weights for the fine-tuned `roberta-base` model (478 MB):
 
 ```bash
-wget https://storage.googleapis.com/gpt-2/detector-models/v1/detector-base.pt
+wget https://openaipublic.azureedge.net/gpt-2/detector-models/v1/detector-base.pt
 ```
 
 or `roberta-large` model (1.5 GB):
 
 ```bash
-wget https://storage.googleapis.com/gpt-2/detector-models/v1/detector-large.pt
+wget https://openaipublic.azureedge.net/gpt-2/detector-models/v1/detector-large.pt
 ```
 
 These RoBERTa-based models are fine-tuned with a mixture of temperature-1 and nucleus sampling outputs,
diff --git a/detector/download.py b/detector/download.py
index 3d4f931..2ed9032 100644
--- a/detector/download.py
+++ b/detector/download.py
@@ -30,7 +30,7 @@ def download(*datasets, data_dir='data'):
             if os.path.isfile(output_file):
                 continue
 
-            r = requests.get("https://storage.googleapis.com/gpt-2/output-dataset/v1/" + filename, stream=True)
+            r = requests.get("https://openaipublic.azureedge.net/gpt-2/output-dataset/v1/" + filename, stream=True)
 
             with open(output_file, 'wb') as f:
                 file_size = int(r.headers["content-length"])
diff --git a/download_dataset.py b/download_dataset.py
index 39a8f8c..fec082b 100644
--- a/download_dataset.py
+++ b/download_dataset.py
@@ -17,7 +17,7 @@ for ds in [
 ]:
     for split in ['train', 'valid', 'test']:
         filename = ds + "." + split + '.jsonl'
-        r = requests.get("https://openaipublic.blob.core.windows.net/gpt-2/output-dataset/v1/" + filename, stream=True)
+        r = requests.get("https://openaipublic.azureedge.net/gpt-2/output-dataset/v1/" + filename, stream=True)
 
         with open(os.path.join(subdir, filename), 'wb') as f:
             file_size = int(r.headers["content-length"])