diff --git a/pandas_gbq/gbq.py b/pandas_gbq/gbq.py
index b04ad131..feffd858 100644
--- a/pandas_gbq/gbq.py
+++ b/pandas_gbq/gbq.py
@@ -1091,6 +1091,14 @@ def to_gbq(
         .. versionadded:: 0.23.3
     """
 
+    # If we get a bigframes.pandas.DataFrame object, it may be possible to use
+    # the code paths here, but it could potentially be quite expensive because
+    # of the queries involved in type detection. It would be safer just to
+    # fail early if there are bigframes-y methods available.
+    # https://github.com/googleapis/python-bigquery-pandas/issues/824
+    if hasattr(dataframe, "to_pandas") and hasattr(dataframe, "to_gbq"):
+        raise TypeError(f"Expected a pandas.DataFrame, but got {repr(type(dataframe))}")
+
     _test_google_api_imports()
 
     from google.api_core import exceptions as google_exceptions
diff --git a/tests/unit/test_to_gbq.py b/tests/unit/test_to_gbq.py
index 60ea8025..f4012dc8 100644
--- a/tests/unit/test_to_gbq.py
+++ b/tests/unit/test_to_gbq.py
@@ -11,6 +11,16 @@
 from pandas_gbq import gbq
 
 
+class FakeDataFrame:
+    """A fake bigframes DataFrame to avoid depending on bigframes."""
+
+    def to_gbq(self):
+        """Fake to_gbq() to mimic a bigframes object."""
+
+    def to_pandas(self):
+        """Fake to_pandas() to mimic a bigframes object."""
+
+
 @pytest.fixture
 def expected_load_method(mock_bigquery_client):
     return mock_bigquery_client.load_table_from_dataframe
@@ -66,6 +76,15 @@ def test_to_gbq_load_method_translates_exception(
     expected_load_method.assert_called_once()
 
 
+def test_to_gbq_with_bigframes_raises_typeerror():
+    dataframe = FakeDataFrame()
+
+    with pytest.raises(
+        TypeError, match=r"Expected a pandas.DataFrame, but got .+FakeDataFrame"
+    ):
+        gbq.to_gbq(dataframe, "my_dataset.my_table", project_id="myproj")
+
+
 def test_to_gbq_with_if_exists_append(mock_bigquery_client, expected_load_method):
     from google.cloud.bigquery import SchemaField