BENGALURU: A day after the Chief Minister’s Office raised concern over auto translation from Kannada to English, Meta (Previously Facebook) acknowledged the glitch, apologised, and claimed to have fixed the issue.
The CM had stated that the faulty auto-translation of Kannada content on Meta platforms is distorting facts and misleading users.
Siddaramaiah in his ‘X’ handle stated that on July 15, he paid his respects to the mortal remains of the late B Saroja Devi, a moment of deep personal and public sorrow. A post about this was shared in Kannada from the official Facebook page of Chief Minister of Karnataka.
“Unfortunately, some people saw a wrong English translation of that post by default on their Facebook feed. The post was written only in Kannada. No English version was posted, and no auto-translation was added or requested from our side,” he stated.
Siddaramaiah pointed out that Facebook’s translation feature shows up on user feeds based on their settings, and content creators have no way to disable or control the auto-translation feature on viewers feed. “In many cases, Facebook shows auto-translated versions of posts by default in user feeds. The original Kannada post remains unchanged and can be viewed by selecting “See original” option’’ he added.
The CM also added that Meta/Facebook has often previously faced similar issues globally for such auto-translation errors, including in Myanmar (2018), Palestine (2017), and most recently in Malaysia (2024) where poor translations led to serious miscommunication and public backlash.
The CM said that Facebook has been in operation for nearly two decades and has a vast user base in India, including a large number of Kannada-speaking users. “Despite this, it is unfortunate that a sound and reliable translation system has not been put in place. We hope this concern is taken seriously.’’ he said.
The CM also urged people not to spread confusion or misinformation based on an error that did not originate from us. “It is time Meta puts in place better and more responsible translation systems,” he added.