diff --git a/badwords/default_words.txt b/badwords/default_words.txt
index fecd33c..1885c06 100644
--- a/badwords/default_words.txt
+++ b/badwords/default_words.txt
@@ -19,6 +19,7 @@ bajingan*
 bangke
 bangs?t*
 bangsad
+bangsat
 bastard*
 batshit
 bedebah
diff --git a/config.py b/config.py
index 3f42db0..65fbb1c 100644
--- a/config.py
+++ b/config.py
@@ -116,7 +116,7 @@
     "Third warning: you will be temporarily banned for {duration} minutes due to channel join spam.",
 ]
 BADWORDS_ENABLED: bool = True
-BADWORDS_INTERCEPT_TYPES: List[str] = ["CHANNEL"]
+BADWORDS_INTERCEPT_TYPES: List[str] = ["CHANNEL", "PRIVATE"]
 BADWORDS_FILE: str = "badwords/words.txt"
 BADWORDS_IGNORE_ADMINS: bool = False
 BADWORDS_PROFILE_CHECK_ENABLED: bool = True
diff --git a/moderation_utils.py b/moderation_utils.py
index 199e4fa..f448f8f 100644
--- a/moderation_utils.py
+++ b/moderation_utils.py
@@ -19,7 +19,7 @@ def check_text_badwords(client, textmessage, content=None):
             content = from_tt_char(textmessage.szMessage)
         if not content:
             return
-        if getattr(config, "BADWORDS_IGNORE_ADMINS", True):
+        if getattr(config, "BADWORDS_IGNORE_ADMINS", False):
             try:
                 u = client.getUser(from_uid)
                 if u.uUserType & UserType.USERTYPE_ADMIN:
@@ -39,7 +39,7 @@ def check_user_profile_badwords(client, user):
     try:
         if user.nUserID == (client.getMyUserID() or 0):
             return
-        if getattr(config, "BADWORDS_IGNORE_ADMINS", True) and (
+        if getattr(config, "BADWORDS_IGNORE_ADMINS", False) and (
             user.uUserType & UserType.USERTYPE_ADMIN
         ):
             return
diff --git a/tests/test_badwords.py b/tests/test_badwords.py
index 28eda37..b651b1b 100644
--- a/tests/test_badwords.py
+++ b/tests/test_badwords.py
@@ -78,6 +78,16 @@ def test_every_entry_is_usable(self):
             with self.subTest(entry=entry):
                 self.assertEqual(checker.pattern_error(entry), "")
 
+    def test_shipped_list_matches_toxic_phrases(self):
+        filter_ = BadWordsFilter()
+        filter_.load_file(os.path.join(ROOT, "badwords", DEFAULT_LIST_NAME))
+        self.assertTrue(filter_.contains("apa lu bangsat"))
+        matches = filter_.matching_entries("apa lu bangsat")
+        self.assertTrue(any("bangsat" in m or "bangs" in m for m in matches))
+        self.assertTrue(filter_.contains("APA LU BANGSAT!"))
+        self.assertTrue(filter_.contains("@ai apa lu bangsat"))
+
+
 
 if __name__ == "__main__":
     unittest.main()