diff --git a/.grit/patterns/python/openai.md b/.grit/patterns/python/openai.md
index 01e1ada9..a19f056e 100644
--- a/.grit/patterns/python/openai.md
+++ b/.grit/patterns/python/openai.md
@@ -198,6 +198,17 @@ pattern pytest_patch() {
     },
 }
 
+
+
+// When there is a variable used by an openai call, make sure it isn't subscripted
+pattern fix_downstream_openai_usage() {
+    $var where {
+        $program <: maybe contains bubble($var) `$x['$y']` as $sub => `$x.$y` where {
+          $sub <: contains $var
+        }
+    }
+}
+
 pattern openai_main($client, $azure) {
     $body where {
         if ($client <: undefined) {
@@ -257,6 +268,9 @@ pattern openai_main($client, $azure) {
           contains `import openai` as $import_stmt where {
               $body <: contains bubble($has_sync, $has_async, $has_openai_import, $body, $client, $azure) `openai.$res.$func($params)` as $stmt where {
                   $res <: rewrite_whole_fn_call(import = $has_openai_import, $has_sync, $has_async, $res, $func, $params, $stmt, $body, $client, $azure),
+                  $stmt <: maybe within bubble($stmt) `$var = $stmt` where {
+                      $var <: fix_downstream_openai_usage()
+                  }
               },
           },
           contains `from openai import $resources` as $partial_import_stmt where {
@@ -562,3 +576,51 @@ response = client.chat.completions.create(
   ]
 )
 ```
+
+## Fix subscripting
+
+The new API does not support subscripting on the outputs.
+
+```python
+import openai
+
+model, token_limit, prompt_cost, comp_cost = 'gpt-4-32k', 32_768, 0.06, 0.12
+
+completion = openai.ChatCompletion.create(
+    model=model,
+    messages=[
+        {"role": "system", "content": system},
+        {"role": "user", "content":
+         user + text},
+    ]
+)
+output = completion['choices'][0]['message']['content']
+
+prom = completion['usage']['prompt_tokens']
+comp = completion['usage']['completion_tokens']
+
+# unrelated variable
+foo = something['else']
+```
+
+```python
+from openai import OpenAI
+
+client = OpenAI()
+
+model, token_limit, prompt_cost, comp_cost = 'gpt-4-32k', 32_768, 0.06, 0.12
+
+completion = client.chat.completions.create(model=model,
+messages=[
+    {"role": "system", "content": system},
+    {"role": "user", "content":
+     user + text},
+])
+output = completion.choices[0].message.content
+
+prom = completion.usage.prompt_tokens
+comp = completion.usage.completion_tokens
+
+# unrelated variable
+foo = something['else']
+```
diff --git a/.grit/patterns/python/openai_azure.md b/.grit/patterns/python/openai_azure.md
index 65a1d22b..1856f2e7 100644
--- a/.grit/patterns/python/openai_azure.md
+++ b/.grit/patterns/python/openai_azure.md
@@ -64,7 +64,7 @@ response = client.chat.completions.create(
   ]
 )
 
-print(response['choices'][0]['message']['content'])
+print(response.choices[0].message.content)
 ```
 
 ## Embeddings
@@ -99,7 +99,7 @@ response = client.embeddings.create(
   input="Your text string goes here",
   model="YOUR_DEPLOYMENT_NAME"
 )
-embeddings = response['data'][0]['embedding']
+embeddings = response.data[0].embedding
 print(embeddings)
 ```