diff --git a/nucliadb/src/nucliadb/search/search/chat/prompt.py b/nucliadb/src/nucliadb/search/search/chat/prompt.py
index b832f21f80..cb8f82b5cd 100644
--- a/nucliadb/src/nucliadb/search/search/chat/prompt.py
+++ b/nucliadb/src/nucliadb/search/search/chat/prompt.py
@@ -217,7 +217,10 @@ async def get_resource_field_extracted_text(
     try:
         field_type, field_key = field_id.strip("/").split("/")
     except ValueError:
-        logger.error(f"Invalid field id: {field_id}. Skipping getting extracted text.")
+        logger.info(
+            f"Invalid field id: {field_id}. Skipping getting extracted text.",
+            extra={"kbid": kb_obj.kbid},
+        )
         return None
     field = await resource.get_field(field_key, KB_REVERSE[field_type], load=False)
     if field is None:
@@ -389,6 +392,7 @@ async def hierarchy_prompt_context(
                 start=0,
                 end=500,
                 extracted_text_cache=etcache,
+                log_on_missing_field=False,
             )
             summary_text = await paragraphs.get_paragraph_text(
                 kbid=kbid,
@@ -397,6 +401,7 @@ async def hierarchy_prompt_context(
                 start=0,
                 end=1000,
                 extracted_text_cache=etcache,
+                log_on_missing_field=False,
             )
             resources[rid] = ExtraCharsParagraph(
                 title=title_text,
diff --git a/nucliadb/src/nucliadb/search/search/paragraphs.py b/nucliadb/src/nucliadb/search/search/paragraphs.py
index b62dce2932..0e1e1754b0 100644
--- a/nucliadb/src/nucliadb/search/search/paragraphs.py
+++ b/nucliadb/src/nucliadb/search/search/paragraphs.py
@@ -117,6 +117,7 @@ async def get_paragraph_from_full_text(
     end: int,
     split: Optional[str] = None,
     extracted_text_cache: Optional[ExtractedTextCache] = None,
+    log_on_missing_field: bool = True,
 ) -> str:
     """
     Pull paragraph from full text stored in database.
@@ -125,13 +126,14 @@ async def get_paragraph_from_full_text(
     """
     extracted_text = await get_field_extracted_text(field, cache=extracted_text_cache)
     if extracted_text is None:
-        logger.warning(
-            "Extracted_text for field does not exist on DB. This should not happen.",
-            extra={
-                "field_id": field.resource_unique_id,
-                "kbid": field.kbid,
-            },
-        )
+        if log_on_missing_field:
+            logger.warning(
+                "Extracted_text for field does not exist on DB. This should not happen.",
+                extra={
+                    "field_id": field.resource_unique_id,
+                    "kbid": field.kbid,
+                },
+            )
         return ""
 
     if split not in (None, ""):
@@ -156,14 +158,16 @@ async def get_paragraph_text(
         ResourceORM
     ] = None,  # allow passing in orm_resource to avoid extra DB calls or txn issues
     extracted_text_cache: Optional[ExtractedTextCache] = None,
+    log_on_missing_field: bool = True,
 ) -> str:
     if orm_resource is None:
         orm_resource = await get_resource_from_cache(kbid, rid)
         if orm_resource is None:
-            logger.warning(
-                "Resource does not exist on DB. This should not happen.",
-                extra={"resource_id": rid, "kbid": kbid, "field": field},
-            )
+            if log_on_missing_field:
+                logger.warning(
+                    "Resource does not exist on DB. This should not happen.",
+                    extra={"resource_id": rid, "kbid": kbid, "field": field},
+                )
             return ""
 
     _, field_type, field = field.split("/")
@@ -176,6 +180,7 @@ async def get_paragraph_text(
         end=end,
         split=split,
         extracted_text_cache=extracted_text_cache,
+        log_on_missing_field=log_on_missing_field,
     )
 
     if highlight: