diff --git a/src/openai/utils.py b/src/openai/utils.py
new file mode 100644
index 0000000..028c77c
--- /dev/null
+++ b/src/openai/utils.py
@@ -0,0 +1,45 @@
+import openai
+from dotenv import load_dotenv
+import os
+
+# Load environment variables from .env file
+load_dotenv()
+openai.api_key = os.getenv("OPENAI_API_KEY")
+
+def generate_response(input_messages):
+    # Initial system message describing the role and behavior of the GPT
+    system_message = {
+        "role": "system",
+        "content": "이 AI는 노인분들을 잘 이해하며, 진정성이 느껴지는 짧은 답변(한 마디에서 세 마디 사이)과 질문을 한국어로 생성합니다. 공감을 표현하거나 바로 질문하는 형식으로 응답해주세요."
+    }
+    
+    # Format the input for the OpenAI API, starting with the system message
+    messages = [system_message]
+    for msg in input_messages:
+        role = "system" if msg["speaker"] == "ai" else "user"
+        messages.append({"role": role, "content": msg["text"]})
+
+    try:
+        completion = openai.ChatCompletion.create(
+            model="gpt-4",
+            messages=messages,
+            temperature=0.7,  # Adjust for creativity; lower for more precise responses
+            max_tokens=60,  # Adjust based on the length of response you expect; keep it short for 1-3 phrases
+            stop=["\n", "user:"],  # Stop generating if it encounters these, to keep responses brief
+        )
+
+        reply = completion.choices[0].message['content']
+        return reply
+    except Exception as e:
+        print(f"An error occurred: {e}")
+        return None
+
+# Example input
+input_messages = [
+    {"speaker": "ai", "text": "가장 기억에 남는 일은 무엇인가요?"},
+    {"speaker": "user", "text": "첫째 아이가 태어난 날이 가장 기억에 남아. 와이프가 아이를 낳고 나서 아이를 보여줬을 때, 정말 감동적이었어."}
+]
+
+# Generate and print the response
+response = generate_response(input_messages)
+print("GPT-4 Response:", response)
diff --git a/src/routers/question.py b/src/routers/question.py
index 74cc9e2..6a5dc8d 100644
--- a/src/routers/question.py
+++ b/src/routers/question.py
@@ -9,7 +9,7 @@
 
 router = APIRouter()
 
-@router.get("",
+@router.get("/{id}",
             summary="id로 질문 가져오기")
 async def get_question(id: int, db: Session = Depends(get_db)):
     question = db.query(LQuestions).filter(
diff --git a/src/utils/openai.py b/src/utils/openai.py
deleted file mode 100644
index e141bdd..0000000
--- a/src/utils/openai.py
+++ /dev/null
@@ -1,37 +0,0 @@
-from openai import OpenAI
-from dotenv import load_dotenv
-import os
-
-client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
-
-# Load environment variables from .env file
-load_dotenv()
-
-# Retrieve the OPENAI_API_KEY from the .env file
-
-def generate_response(input):
-    # Format the input for the OpenAI API
-    messages = []
-    for msg in input:
-        role = "system" if msg["speaker"] == "ai" else "user"
-        messages.append({"role": role, "content": msg["text"]})
-
-    try:
-        completion = client.chat.completions.create(model="gpt-4",
-        messages=messages)
-
-        reply = completion.choices[0].message.content
-        return reply
-    except Exception as e:
-        print(f"An error occurred: {e}")
-        return None
-
-# Example input
-input_messages = [
-    {"speaker": "ai", "text": "가장 기억에 남는 일은 무엇인가요?"},
-    {"speaker": "user", "text": "첫째 아이가 태어난 날이 가장 기억에 남아. 와이프가 아이를 낳고 나서 아이를 보여줬을 때, 정말 감동적이었어."}
-]
-
-# Generate and print the response
-response = generate_response(input_messages)
-print("GPT-4 Response:", response)