diff --git a/src/runtime/lexer.c b/src/runtime/lexer.c
index 123a29fd..5646e101 100644
--- a/src/runtime/lexer.c
+++ b/src/runtime/lexer.c
@@ -36,11 +36,17 @@ static void ts_lexer__get_lookahead(Lexer *self) {
   const uint8_t *chunk = (const uint8_t *)self->chunk + position_in_chunk;
   uint32_t size = self->chunk_size - position_in_chunk + 1;
 
-  if (self->input.encoding == TSInputEncodingUTF8)
-    self->lookahead_size =
-      utf8proc_iterate(chunk, size, &self->data.lookahead);
-  else
+  if (self->input.encoding == TSInputEncodingUTF8) {
+    int64_t lookahead_size = utf8proc_iterate(chunk, size, &self->data.lookahead);
+    if (lookahead_size < 0) {
+      self->lookahead_size = 1;
+    } else {
+      self->lookahead_size = lookahead_size;
+    }
+  }
+  else {
     self->lookahead_size = utf16_iterate(chunk, size, &self->data.lookahead);
+  }
 }
 
 static void ts_lexer__advance(void *payload, bool skip) {
diff --git a/test/runtime/parser_test.cc b/test/runtime/parser_test.cc
index 7dfcf26b..7e409d45 100644
--- a/test/runtime/parser_test.cc
+++ b/test/runtime/parser_test.cc
@@ -473,6 +473,16 @@ describe("Parser", [&]() {
       AssertThat(ts_node_end_char(root), Equals(strlen("'OOO - DD';")));
       AssertThat(ts_node_end_byte(root), Equals(strlen("'\u03A9\u03A9\u03A9 \u2014 \u0394\u0394';")));
     });
+
+    it("handles non-UTF8 characters", [&]() {
+      // ts_document_set_logger(document, stderr_logger_new(true));
+      ts_document_print_debugging_graphs(document, true);
+      ts_document_set_language(document, load_real_language("javascript"));
+      ts_document_set_input_string(document, "cons\xeb\x00e=ls\x83l6hi');\x0a");
+      ts_document_parse(document);
+
+      AssertThat(ts_node_end_byte(root), Equals(strlen("cons\xeb\x00e=ls\x83l6hi');\x0a")));
+    });
   });
 });