Naver will automatically disable comments on articles if malicious comments exceed a set threshold.
The company said it will begin enforcing the automatic comment-disable policy on April 23 to promote a healthier news environment.
Naver uses a system called CleanBot to detect malicious comments across all news sections in real time.
When the proportion of malicious comments reaches a predefined level, Naver sends an advance notice that reads, "Malicious comments in this article are increasing and comments may be automatically closed."
If malicious comments surpass the threshold, the comment box is replaced with a Green Internet campaign banner and a notice: "CleanBot has detected a large number of malicious comments, and the comment service is not available."
Articles that have their comments disabled for exceeding the malicious-comment threshold cannot be reopened, nor can their comment settings be changed.
CleanBot was introduced in 2019, and Naver has steadily improved its ability to detect profanity, sexually explicit language, hate speech, and derogatory expressions.
Naver said it plans to upgrade the AI CleanBot model by the end of the month.
Kim Su-hyang, a Naver leader, said, "Naver has been enhancing CleanBot and, together with last month's move to stop offering comments under political and election articles, is working to make comment sections healthier spaces for discussion."
