diff --git a/references/go/data-handling.md b/references/go/data-handling.md
index 18ccf57..7afc035 100644
--- a/references/go/data-handling.md
+++ b/references/go/data-handling.md
@@ -255,6 +255,27 @@ err := workflow.UpsertMemo(ctx, map[string]interface{}{
 })
 ```
 
+### Memo encoding: default vs. user `DataConverter`
+
+By default, Memo values are serialized with the **default** Data Converter — even when `client.Options{DataConverter}` is set to a custom converter.
+That means a `CodecDataConverter` you configured to encrypt Workflow inputs/results does **not** apply to Memo fields by default.
+
+Go SDK v1.41.0 added the opt-in SDK flag `SDKFlagMemoUserDCEncode` (numeric ID `7`).
+When enabled, `StartWorkflow`, `SignalWithStartWorkflow`, `ScheduleClient.Create`, `workflow.UpsertMemo`, and `workflow.ExecuteChildWorkflow` route Memo values through the user `DataConverter` from `client.Options`.
+On encode error, the SDK falls back to the default Data Converter; if the default also errors, the **user converter's** error is returned.
+
+The flag is **disabled by default** in `sdkFlagsAllowed` in v1.41.0 through v1.43.0.
+Enable it at process start by setting the env var `TEMPORAL_SDK_FLAG_7=1`.
+There is no public `client.Options` or `worker.Options` field for this — the env var is the only knob.
+
+To encrypt Memos with a `PayloadCodec`, both must be true: (a) the codec is part of the `DataConverter` passed to `client.Dial` (e.g., via `converter.NewCodecDataConverter`), and (b) `TEMPORAL_SDK_FLAG_7=1` is set on every client process and worker process that writes Memos.
+
+Scope notes:
+
+- The flag affects Memo encoding only; Search Attributes use a separate typed-value path.
+- For in-Workflow calls (`UpsertMemo`, `ExecuteChildWorkflow`), the flag is gated through `TryUse`, so the decision is recorded in history and replay stays deterministic.
+- Memos remain "non-indexed metadata" with no type safety and eventual consistency — enabling the flag does not change those properties.
+
 ## Best Practices
 
 1. Use structs with exported fields for inputs and outputs
@@ -262,3 +283,4 @@ err := workflow.UpsertMemo(ctx, map[string]interface{}{
 3. Keep payloads small -- see `references/core/gotchas.md` for limits
 4. Use `PayloadCodec` for encryption; never store sensitive data unencrypted
 5. Configure the same data converter on both client and worker
+6. Don't put sensitive data in a Memo unless every process that writes memos sets `TEMPORAL_SDK_FLAG_7=1` **and** the client's `DataConverter` includes the encrypting `PayloadCodec` — otherwise memos bypass the codec
diff --git a/references/java/integrations/spring-ai.md b/references/java/integrations/spring-ai.md
index 5ee0704..ae5154f 100644
--- a/references/java/integrations/spring-ai.md
+++ b/references/java/integrations/spring-ai.md
@@ -217,7 +217,6 @@ Media image = new Media(MimeTypeUtils.IMAGE_PNG, URI.create("https://cdn.example
 
 For anything larger than a small thumbnail, route the bytes to a binary store from an Activity and pass only the URL across the conversation.
 
-
 ## Vector stores, embeddings, and MCP
 
 When the corresponding Spring AI modules (`spring-ai-rag`, `spring-ai-mcp`) are on the classpath, the integration registers Activities for vector stores, embeddings, and MCP tool calls automatically. Inject the matching Spring AI types into your Activities or Workflows and use them as you would in any Spring AI application — each operation executes through a Temporal Activity.