diff --git a/google/cloud/retail/v2alpha/BUILD.bazel b/google/cloud/retail/v2alpha/BUILD.bazel
index 716511a30d..366db163d8 100644
--- a/google/cloud/retail/v2alpha/BUILD.bazel
+++ b/google/cloud/retail/v2alpha/BUILD.bazel
@@ -49,6 +49,7 @@ proto_library(
         "project_service.proto",
         "promotion.proto",
         "purge_config.proto",
+        "safety.proto",
         "search_service.proto",
         "serving_config.proto",
         "serving_config_service.proto",
diff --git a/google/cloud/retail/v2alpha/conversational_search_service.proto b/google/cloud/retail/v2alpha/conversational_search_service.proto
index 21e33ba6c4..f06c141453 100644
--- a/google/cloud/retail/v2alpha/conversational_search_service.proto
+++ b/google/cloud/retail/v2alpha/conversational_search_service.proto
@@ -21,6 +21,7 @@ import "google/api/client.proto";
 import "google/api/field_behavior.proto";
 import "google/api/resource.proto";
 import "google/cloud/retail/v2alpha/common.proto";
+import "google/cloud/retail/v2alpha/safety.proto";
 import "google/cloud/retail/v2alpha/search_service.proto";
 
 option csharp_namespace = "Google.Cloud.Retail.V2Alpha";
@@ -126,6 +127,12 @@ message ConversationalSearchRequest {
       // Default value.
       MODE_UNSPECIFIED = 0;
 
+      // Disable Conversational Filtering.
+      DISABLED = 1;
+
+      // Enabled Conversational Filtering with default Conversational Search.
+      ENABLED = 2;
+
       // Enabled Conversational Filtering without default Conversational Search.
       CONVERSATIONAL_FILTER_ONLY = 3;
     }
@@ -216,6 +223,30 @@ message ConversationalSearchRequest {
   // parameters.
   ConversationalFilteringSpec conversational_filtering_spec = 8
       [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. The user labels applied to a resource must meet the following
+  // requirements:
+  //
+  // * Each resource can have multiple labels, up to a maximum of 64.
+  // * Each label must be a key-value pair.
+  // * Keys have a minimum length of 1 character and a maximum length of 63
+  //   characters and cannot be empty. Values can be empty and have a maximum
+  //   length of 63 characters.
+  // * Keys and values can contain only lowercase letters, numeric characters,
+  //   underscores, and dashes. All characters must use UTF-8 encoding, and
+  //   international characters are allowed.
+  // * The key portion of a label must be unique. However, you can use the same
+  //   key with multiple resources.
+  // * Keys must start with a lowercase letter or international character.
+  //
+  // See [Google Cloud
+  // Document](https://cloud.google.com/resource-manager/docs/creating-managing-labels#requirements)
+  // for more details.
+  map<string, string> user_labels = 12 [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. The safety settings to be applied to the generated content.
+  repeated SafetySetting safety_settings = 14
+      [(google.api.field_behavior) = OPTIONAL];
 }
 
 // Response message for
@@ -269,6 +300,46 @@ message ConversationalSearchResponse {
     AdditionalFilter additional_filter = 2;
   }
 
+  // The state of the response generation.
+  enum State {
+    // Unknown.
+    STATE_UNSPECIFIED = 0;
+
+    // Response generation is being streamed.
+    STREAMING = 1;
+
+    // Response generation has succeeded.
+    SUCCEEDED = 2;
+  }
+
+  // The types Retail classifies the search query as.
+  //
+  // Supported values are:
+  //
+  // - "ADVERSARIAL"
+  // - "CHITCHAT"
+  // - "JAILBREAK"
+  // - "ORDER_SUPPORT"
+  // - "SIMPLE_PRODUCT_SEARCH"
+  // - "INTENT_REFINEMENT"
+  // - "PRODUCT_DETAILS"
+  // - "PRODUCT_COMPARISON"
+  // - "DEALS_AND_COUPONS"
+  // - "STORE_RELEVANT"
+  // - "BLOCKLISTED"
+  // - "BEST_PRODUCT"
+  // - "RETAIL_SUPPORT"
+  // - "DISABLED"
+  // clang-format off
+  // clang-format on
+  repeated string user_query_types = 10;
+
+  // The conversational answer-based text response generated by the Server.
+  string conversational_text_response = 2;
+
+  // The conversational followup question generated for Intent refinement.
+  FollowupQuestion followup_question = 3;
+
   // Conversation UUID. This field will be stored in client side storage to
   // maintain the conversation session with server and will be used for next
   // search request's
@@ -284,4 +355,7 @@ message ConversationalSearchResponse {
   // This field specifies all related information that is needed on client
   // side for UI rendering of conversational filtering search.
   ConversationalFilteringResult conversational_filtering_result = 7;
+
+  // Output only. The state of the response generation.
+  State state = 9 [(google.api.field_behavior) = OUTPUT_ONLY];
 }
diff --git a/google/cloud/retail/v2alpha/safety.proto b/google/cloud/retail/v2alpha/safety.proto
new file mode 100644
index 0000000000..636fdace8e
--- /dev/null
+++ b/google/cloud/retail/v2alpha/safety.proto
@@ -0,0 +1,95 @@
+// Copyright 2025 Google LLC
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+syntax = "proto3";
+
+package google.cloud.retail.v2alpha;
+
+import "google/api/field_behavior.proto";
+
+option csharp_namespace = "Google.Cloud.Retail.V2Alpha";
+option go_package = "cloud.google.com/go/retail/apiv2alpha/retailpb;retailpb";
+option java_multiple_files = true;
+option java_outer_classname = "SafetyProto";
+option java_package = "com.google.cloud.retail.v2alpha";
+option objc_class_prefix = "RETAIL";
+option php_namespace = "Google\\Cloud\\Retail\\V2alpha";
+option ruby_package = "Google::Cloud::Retail::V2alpha";
+
+// Harm categories that will block the content.
+enum HarmCategory {
+  // The harm category is unspecified.
+  HARM_CATEGORY_UNSPECIFIED = 0;
+
+  // The harm category is hate speech.
+  HARM_CATEGORY_HATE_SPEECH = 1;
+
+  // The harm category is dangerous content.
+  HARM_CATEGORY_DANGEROUS_CONTENT = 2;
+
+  // The harm category is harassment.
+  HARM_CATEGORY_HARASSMENT = 3;
+
+  // The harm category is sexually explicit content.
+  HARM_CATEGORY_SEXUALLY_EXPLICIT = 4;
+
+  // The harm category is civic integrity.
+  HARM_CATEGORY_CIVIC_INTEGRITY = 5;
+}
+
+// Safety settings.
+message SafetySetting {
+  // Probability based thresholds levels for blocking.
+  enum HarmBlockThreshold {
+    // Unspecified harm block threshold.
+    HARM_BLOCK_THRESHOLD_UNSPECIFIED = 0;
+
+    // Block low threshold and above (i.e. block more).
+    BLOCK_LOW_AND_ABOVE = 1;
+
+    // Block medium threshold and above.
+    BLOCK_MEDIUM_AND_ABOVE = 2;
+
+    // Block only high threshold (i.e. block less).
+    BLOCK_ONLY_HIGH = 3;
+
+    // Block none.
+    BLOCK_NONE = 4;
+
+    // Turn off the safety filter.
+    OFF = 5;
+  }
+
+  // Probability vs severity.
+  enum HarmBlockMethod {
+    // The harm block method is unspecified.
+    HARM_BLOCK_METHOD_UNSPECIFIED = 0;
+
+    // The harm block method uses both probability and severity scores.
+    SEVERITY = 1;
+
+    // The harm block method uses the probability score.
+    PROBABILITY = 2;
+  }
+
+  // Harm category.
+  HarmCategory category = 1;
+
+  // The harm block threshold.
+  HarmBlockThreshold threshold = 2;
+
+  // Optional. Specify if the threshold is used for probability or severity
+  // score. If not specified, the threshold is used for probability score.
+  HarmBlockMethod method = 3 [(google.api.field_behavior) = OPTIONAL];
+}