diff --git a/PlotLine.Tests/Program.cs b/PlotLine.Tests/Program.cs
index 6d81e95..0c28629 100644
--- a/PlotLine.Tests/Program.cs
+++ b/PlotLine.Tests/Program.cs
@@ -39,6 +39,7 @@ var tests = new (string Name, Action Test)[]
     ("Illustration starter generation plan skips successful items", IllustrationStarterGenerationPlanSkipsSuccessfulItems),
     ("Story Intelligence prototype uses starter illustration codes", StoryIntelligencePrototypeUsesStarterIllustrationCodes),
     ("Story Intelligence simulation mode remains available", StoryIntelligenceSimulationModeRemainsAvailable),
+    ("Story Intelligence does not retry exhausted billing quota", StoryIntelligenceDoesNotRetryExhaustedBillingQuota),
     ("Story Intelligence visualisation contract omits raw manuscript text", StoryIntelligenceVisualisationContractOmitsRawManuscriptText),
     ("Illustration bulk retry result reports counts", IllustrationBulkRetryResultReportsCounts),
     ("Illustration bulk retry skips blocking duplicate stable codes", IllustrationBulkRetrySkipsBlockingDuplicateStableCodes),
@@ -698,6 +699,14 @@ static void StoryIntelligenceSimulationModeRemainsAvailable()
     Assert(prototype.Scenes.Any(scene => scene.PovCharacterId == "rosie"), "Simulation should still exercise POV transitions.");
 }
 
+static void StoryIntelligenceDoesNotRetryExhaustedBillingQuota()
+{
+    var client = File.ReadAllText(Path.Combine(AppContext.BaseDirectory, "../../../../PlotLine/Services/StoryIntelligenceOpenAIInfrastructure.cs"));
+
+    Assert(client.Contains("IsInsufficientQuota(responseText)", StringComparison.Ordinal), "OpenAI insufficient_quota responses must be detected before retrying.");
+    Assert(client.Contains("request was not retried", StringComparison.Ordinal), "OpenAI insufficient_quota responses should fail clearly without retry churn.");
+}
+
 static void StoryIntelligenceVisualisationContractOmitsRawManuscriptText()
 {
     var model = new StoryIntelligenceExperiencePrototypeViewModel
diff --git a/PlotLine/Services/StoryIntelligenceOpenAIInfrastructure.cs b/PlotLine/Services/StoryIntelligenceOpenAIInfrastructure.cs
index a78a5cb..218710b 100644
--- a/PlotLine/Services/StoryIntelligenceOpenAIInfrastructure.cs
+++ b/PlotLine/Services/StoryIntelligenceOpenAIInfrastructure.cs
@@ -279,6 +279,11 @@ public sealed class StoryIntelligenceClient(
                     return (responseText, retryCount);
                 }
 
+                if (IsInsufficientQuota(responseText))
+                {
+                    throw new InvalidOperationException($"OpenAI billing quota exhausted; request was not retried: {TrimForLog(responseText)}");
+                }
+
                 if (!ShouldRetry(response.StatusCode) || attempt == maxAttempts)
                 {
                     throw new InvalidOperationException($"OpenAI request failed with {(int)response.StatusCode} {response.StatusCode}: {TrimForLog(responseText)}");
@@ -318,6 +323,10 @@ public sealed class StoryIntelligenceClient(
             or HttpStatusCode.ServiceUnavailable
             or HttpStatusCode.GatewayTimeout;
 
+    internal static bool IsInsufficientQuota(string responseText)
+        => responseText.Contains("\"insufficient_quota\"", StringComparison.OrdinalIgnoreCase)
+            || responseText.Contains("exceeded your current quota", StringComparison.OrdinalIgnoreCase);
+
     private static Task DelayForRetryAsync(int attempt, CancellationToken cancellationToken)
         => Task.Delay(TimeSpan.FromSeconds(Math.Pow(2, attempt - 1)), cancellationToken);