diff --git a/lldb/source/Commands/CommandObjectMultiword.cpp b/lldb/source/Commands/CommandObjectMultiword.cpp
--- a/lldb/source/Commands/CommandObjectMultiword.cpp
+++ b/lldb/source/Commands/CommandObjectMultiword.cpp
@@ -274,10 +274,10 @@
 
   StringList new_matches;
   CommandObject *sub_command_object = GetSubcommandObject(arg0, &new_matches);
-  if (sub_command_object == nullptr) {
-    request.AddCompletions(new_matches);
+
+  // The subcommand is ambiguous. The completion isn't meaningful.
+  if (!sub_command_object)
     return;
-  }
 
   // Remove the one match that we got from calling GetSubcommandObject.
   new_matches.DeleteStringAtIndex(0);
diff --git a/lldb/test/API/functionalities/completion/TestCompletion.py b/lldb/test/API/functionalities/completion/TestCompletion.py
--- a/lldb/test/API/functionalities/completion/TestCompletion.py
+++ b/lldb/test/API/functionalities/completion/TestCompletion.py
@@ -880,3 +880,11 @@
         self.complete_from_to("breakpoint set -N n", "breakpoint set -N n")
         self.assertTrue(bp1.AddNameWithErrorHandling("nn"))
         self.complete_from_to("breakpoint set -N ", "breakpoint set -N nn")
+
+    def test_ambiguous_command(self):
+        """Test completing an ambiguous commands"""
+        self.complete_from_to("settings s", ['set', 'show'])
+
+    def test_ambiguous_subcommand(self):
+        """Test completing a subcommand of an ambiguous command"""
+        self.complete_from_to("settings s ta", [])