diff --git a/CHANGELOG.md b/CHANGELOG.md
index a43426fe..dad2e86a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -11,8 +11,8 @@
 - `haiku-rag search`, `ask`, `analyze` and `chat` can cover a configured database
   set. Commands that access one database select it with `--db-name NAME` or
   `--db PATH`.
-- Citation resolution rejects chunk IDs shared by multiple selected databases
-  with `AmbiguousCitationError`.
+- Citation resolution rejects a chunk ID retrieved from, or previously cited
+  from, more than one selected database with `AmbiguousCitationError`.
 
 ### Fixed
 
diff --git a/docs/configuration/storage.md b/docs/configuration/storage.md
index d3396877..9587a10a 100644
--- a/docs/configuration/storage.md
+++ b/docs/configuration/storage.md
@@ -260,10 +260,17 @@ on every result.
 #### Duplicate IDs
 
 IDs are unique within a database, not across databases. Copies of a database
-therefore retain the same IDs. Citation resolution raises
-`AmbiguousCitationError` when a cited chunk ID exists in more than one selected
-database; a shared ID that nothing cites is ignored. The analysis sandbox
-rejects shared document IDs because its mount path is `/documents/{id}/`.
+therefore retain the same IDs.
+
+Citation ambiguity is evaluated against evidence available to the run. A cited
+chunk ID is rejected with `AmbiguousCitationError` if search returned it from
+multiple databases, or it was previously cited from another database. If only
+one retrieved result has the ID, that result is cited. For an ID absent from
+search results, the fallback checks every selected database and rejects
+multiple holders. A shared ID that nothing cites is ignored.
+
+The analysis sandbox rejects shared document IDs because its mount path is
+`/documents/{id}/`.
 
 The chat document filter selects by document ID and applies `id IN (...)` to
 every covered database, so selecting an ID that copies share matches the
diff --git a/tests/multi_db/test_citations.py b/tests/multi_db/test_citations.py
index 25188232..b538b92a 100644
--- a/tests/multi_db/test_citations.py
+++ b/tests/multi_db/test_citations.py
@@ -232,6 +232,41 @@ class TestSharedChunkIds:
         with pytest.raises(ModelRetry, match="another database"):
             await capability._cite(["c1"])
 
+    @pytest.mark.asyncio
+    async def test_one_retrieved_copy_of_a_shared_id_cites_where_it_came_from(
+        self, tmp_path
+    ):
+        """Rejection is about evidence the run holds, not about what the other
+        databases contain. A copy the search never returned did not ground the
+        answer, so the retrieved one is the citation and its database is not a
+        guess. A later question that does retrieve the twin is refused by the
+        citation index."""
+        from tests.capabilities.test_capabilities import Deps, make_context
+
+        config = _config(tmp_path, ["alpha", "beta"])
+        capability = create_capability(config=config, defer_loading=False)
+        deps = Deps(state={"rag": RAGState().model_dump(mode="json")})
+        run = await capability.for_run(make_context(deps))
+        assert run.state is not None
+        # What a run holds when fusion returned one copy of a shared id and
+        # truncated the other: the run never saw the twin.
+        run.state.searches["cats"] = [
+            SearchResult(
+                content="alpha body",
+                score=0.9,
+                source="alpha",
+                chunk_id="c1",
+                document_id="d1",
+                document_uri="test://alpha/one",
+            )
+        ]
+
+        await run._cite(["c1"])
+
+        [cited] = run.state.citation_index.values()
+        assert cited.source == "alpha"
+        assert cited.content == "alpha body"
+
 
 class TestCitationSource:
     def test_a_citation_carries_the_result_source(self):