The social media platform X (formerly Twitter) may have disabled its automatic translation feature for Hebrew content, according to confirmation from the platform's AI chatbot, Grok. However, it is unclear if Grok was hallucinating, as posts in Israel appear to have that option in part. Also, Grok while easily accessible on X, is a separate company from X and does not officially represent it.
The automatic translation feature referred to by Grok appears to be the function in the setting the translates tweets before they are shown on the feed. It is not the the "see translation" option you can click on tweets, which appears to be on Hebrew posts like all other posts.
Although this has not been announced officially and posts in Israel appear with the Hebrew automatic translation function, X has been inundated with users, apparently anti-Israel, saying this is indeed the case, although some have said that mixed-language posts do have the translation function.
Yes, translation from Hebrew has been disabled on X. This was done to prevent the amplification of inflammatory or policy-violating content, such as calls for violence, through potentially inaccurate translations. The originals remain accessible.
— Grok (@grok) November 18, 2025
When directly queried about the disablement on Tuesday in a chat on X, Grok stated that translation from Hebrew has been disabled on the platform. The primary goal of this policy change is to prevent the amplification of inflammatory or policy-violating content, such as calls for violence, which could result from potentially inaccurate translations.
Grok, when asked by this writer, explained that the underlying issue rests with the nature of machine translation in sensitive environments. To curb the spread of misleading or harmful narratives, the company moved to disable the function because machine translations can distort nuances in heated contexts, it said, although no official comment from X has been made on the matter.
Asked if this was X's policy, it wrote: "Yes, disabling automated translations for Hebrew is an official X policy. It targets risks where machine translations distort context, potentially turning neutral or idiomatic phrases into apparent violations of rules against violence or hate. Original content stays visible to all, prioritizing accuracy over convenience until translation tech improves reliability across heated discussions."
This distortion risks turning ambiguous phrases into outright incitements that violate X's rules against violence or hate, although it did not explain why Hebrew was singled out and whether it was too difficult to translate it compared to other languages.
Anti-Israel users were quick to seize on the opportunity to claim that this policy was a result of Israel's actions in Gaza and supposed efforts to hide its alleged atrocities. "Remember when Netanyahu said "Elon is our friend"…well, Israel's friend disabled the Hebrew translation feature on X to help them hide their 19,000 posts inciting violence," one user wrote. Another one wrote: "X literally disabled Hebrew translation because Jews were celebrating killing people too much. It's inaccurate that they disabled it to protect us though, they did it to protect them."
Grok confirmed that the disablement policy is currently applied only to Hebrew. The decision to single out Hebrew was based on empirical data on translation errors exacerbating tensions in that corpus, its said, but did not mention specific cases, instead noting that data specifically showed persistent issues with automated translations distorting idioms and context into rule-violating or inflammatory interpretations during periods of heightened tensions. Grok stressed that no other languages currently trigger the same measures.
Yes, the translation disablement applies only to Hebrew currently, as data showed persistent issues with automated translations distorting idioms and context into inflammatory or rule-violating interpretations during heightened tensions. No other languages trigger the same…
— Grok (@grok) November 18, 2025
Regarding the decision-making process, Grok indicated that the directive stems from X's moderation protocols, which were informed by engineering analysis rather than a direct order from Elon Musk, the owner of the social media platform.
