diff --git a/app/MindWork AI Studio/Assistants/AssistantBase.razor.cs b/app/MindWork AI Studio/Assistants/AssistantBase.razor.cs
index f6582c39..fc9fcfff 100644
--- a/app/MindWork AI Studio/Assistants/AssistantBase.razor.cs	
+++ b/app/MindWork AI Studio/Assistants/AssistantBase.razor.cs	
@@ -293,7 +293,7 @@ public abstract partial class AssistantBase : ComponentBase, IMessageBusReceiver
         // Use the selected provider to get the AI response.
         // By awaiting this line, we wait for the entire
         // content to be streamed.
-        await aiText.CreateFromProviderAsync(this.providerSettings.CreateProvider(this.Logger), this.providerSettings.Model, this.lastUserPrompt, this.chatThread);
+        this.chatThread = await aiText.CreateFromProviderAsync(this.providerSettings.CreateProvider(this.Logger), this.providerSettings.Model, this.lastUserPrompt, this.chatThread);
         
         this.isProcessing = false;
         this.StateHasChanged();
diff --git a/app/MindWork AI Studio/Chat/ContentImage.cs b/app/MindWork AI Studio/Chat/ContentImage.cs
index 9f09e87c..c7e785eb 100644
--- a/app/MindWork AI Studio/Chat/ContentImage.cs	
+++ b/app/MindWork AI Studio/Chat/ContentImage.cs	
@@ -28,7 +28,7 @@ public sealed class ContentImage : IContent, IImageSource
     public Func<Task> StreamingEvent { get; set; } = () => Task.CompletedTask;
 
     /// <inheritdoc />
-    public Task CreateFromProviderAsync(IProvider provider, Model chatModel, IContent? lastPrompt, ChatThread? chatChatThread, CancellationToken token = default)
+    public Task<ChatThread> CreateFromProviderAsync(IProvider provider, Model chatModel, IContent? lastPrompt, ChatThread? chatChatThread, CancellationToken token = default)
     {
         throw new NotImplementedException();
     }
diff --git a/app/MindWork AI Studio/Chat/ContentText.cs b/app/MindWork AI Studio/Chat/ContentText.cs
index 82b5334c..0dccd262 100644
--- a/app/MindWork AI Studio/Chat/ContentText.cs	
+++ b/app/MindWork AI Studio/Chat/ContentText.cs	
@@ -36,10 +36,10 @@ public sealed class ContentText : IContent
     public Func<Task> StreamingEvent { get; set; } = () => Task.CompletedTask;
 
     /// <inheritdoc />
-    public async Task CreateFromProviderAsync(IProvider provider, Model chatModel, IContent? lastPrompt, ChatThread? chatThread, CancellationToken token = default)
+    public async Task<ChatThread> CreateFromProviderAsync(IProvider provider, Model chatModel, IContent? lastPrompt, ChatThread? chatThread, CancellationToken token = default)
     {
         if(chatThread is null)
-            return;
+            return new();
         
         // Call the RAG process. Right now, we only have one RAG process:
         if (lastPrompt is not null)
@@ -115,6 +115,7 @@ public sealed class ContentText : IContent
         
         // Inform the UI that the streaming is done:
         await this.StreamingDone();
+        return chatThread;
     }
 
     #endregion
diff --git a/app/MindWork AI Studio/Chat/IContent.cs b/app/MindWork AI Studio/Chat/IContent.cs
index 8ca94025..be3bf097 100644
--- a/app/MindWork AI Studio/Chat/IContent.cs	
+++ b/app/MindWork AI Studio/Chat/IContent.cs	
@@ -41,7 +41,7 @@ public interface IContent
     /// <summary>
     /// Uses the provider to create the content.
     /// </summary>
-    public Task CreateFromProviderAsync(IProvider provider, Model chatModel, IContent? lastPrompt, ChatThread? chatChatThread, CancellationToken token = default);
+    public Task<ChatThread> CreateFromProviderAsync(IProvider provider, Model chatModel, IContent? lastPrompt, ChatThread? chatChatThread, CancellationToken token = default);
 
     /// <summary>
     /// Returns the corresponding ERI content type.
diff --git a/app/MindWork AI Studio/Components/ChatComponent.razor.cs b/app/MindWork AI Studio/Components/ChatComponent.razor.cs
index 67d805d2..2e05c0fe 100644
--- a/app/MindWork AI Studio/Components/ChatComponent.razor.cs	
+++ b/app/MindWork AI Studio/Components/ChatComponent.razor.cs	
@@ -475,7 +475,7 @@ public partial class ChatComponent : MSGComponentBase, IAsyncDisposable
             // Use the selected provider to get the AI response.
             // By awaiting this line, we wait for the entire
             // content to be streamed.
-            await aiText.CreateFromProviderAsync(this.Provider.CreateProvider(this.Logger), this.Provider.Model, lastUserPrompt, this.ChatThread, this.cancellationTokenSource.Token);
+            this.ChatThread = await aiText.CreateFromProviderAsync(this.Provider.CreateProvider(this.Logger), this.Provider.Model, lastUserPrompt, this.ChatThread, this.cancellationTokenSource.Token);
         }
         
         this.cancellationTokenSource = null;
diff --git a/app/MindWork AI Studio/Pages/Writer.razor.cs b/app/MindWork AI Studio/Pages/Writer.razor.cs
index 76409c32..42b23abc 100644
--- a/app/MindWork AI Studio/Pages/Writer.razor.cs	
+++ b/app/MindWork AI Studio/Pages/Writer.razor.cs	
@@ -139,7 +139,7 @@ public partial class Writer : MSGComponentBase, IAsyncDisposable
         this.isStreaming = true;
         this.StateHasChanged();
         
-        await aiText.CreateFromProviderAsync(this.providerSettings.CreateProvider(this.Logger), this.providerSettings.Model, lastUserPrompt, this.chatThread);
+        this.chatThread = await aiText.CreateFromProviderAsync(this.providerSettings.CreateProvider(this.Logger), this.providerSettings.Model, lastUserPrompt, this.chatThread);
         this.suggestion = aiText.Text;
         
         this.isStreaming = false;
diff --git a/app/MindWork AI Studio/wwwroot/changelog/v0.9.32.md b/app/MindWork AI Studio/wwwroot/changelog/v0.9.32.md
index 3e78d53a..9871b058 100644
--- a/app/MindWork AI Studio/wwwroot/changelog/v0.9.32.md	
+++ b/app/MindWork AI Studio/wwwroot/changelog/v0.9.32.md	
@@ -4,4 +4,5 @@
 - Improved the ERI client to raise an error when the server responds with additional JSON data that is not expected.
 - Improved the error handling in the ERI data source info dialog in cases where servers respond with an invalid message.
 - Improved the error handling for the entire RAG process.
+- Improved chat thread persistence after modifications through the RAG process.
 - Fixed the chat thread we use for the data retrieval by removing the last block, which is meant to be for the final AI answer.
\ No newline at end of file