Insight

How far can AI and our own language tools be trusted?

We test our own correction engine and various outside AI models against Indonesian and regional languages, then publish the numbers as they are. On this page, we sum up each finding in a single picture. The full method and raw data are one click away on every card.

2 findingsEvery finding links to its technical version
FilteredJavaneseClear filter

2 findings

BenchmarkEmpirical
Claude Haiku 4.5
72.22
Claude Sonnet 4.6
91.67
Correct answers out of 36 Javanese politeness-level questions

Tested on Javanese Speech Levels, Anthropic's Cheapest AI Got It Wrong 10 Times

Claude Haiku, Anthropic's cheapest model, answered 10 of 36 Javanese politeness-level questions wrong. Claude Sonnet, its expensive counterpart, missed only 3.

August 19, 2026 · 3 menit bacaRead the finding →
JavaneseEmpirical
Voided, book vs speaker 6Kept in the set 34
Out of the 40 Javanese items we built

6 of 40 Javanese Test Items Voided Because We Trusted the Book Too Much

We built Javanese test items from grammar-book rules. Six of 40 items were dropped, not because AI answered wrong, but because the form the book called wrong turned out to be widely accepted by native speakers.

August 14, 2026 · 3 menit bacaRead the finding →