Naver Upgrades AI Cleanbot 3.0 to Block Malicious Comments

Focus on Blocking Expressions That Trivialize Life and Inflict Secondary Harm

News|
|
By Lee Jin-seok
||
Screenshot of Naver's "AI Cleanbot 3.0." Photo courtesy of Naver - Seoul Economic Daily Technology News from South Korea
Screenshot of Naver's "AI Cleanbot 3.0." Photo courtesy of Naver

Naver (035420.KS) announced Wednesday that it has completed the upgrade of its malicious comment detection system, "AI Cleanbot 3.0." The upgrade intensively blocks comments posted on Naver that promote disregard for life in connection with suicide, death or bodily harm, as well as hateful comments directed at victims of incidents and accidents and their bereaved families.

The "AI Cleanbot 3.0" model is characterized by enhanced malicious comment detection capabilities combined with article context. It has undergone continuous optimization to detect malicious comments by comprehensively analyzing not only the content of news comments but also article titles and body text.

"AI Cleanbot," which Naver introduced in 2019 as the industry's first such system, has expanded its detection scope and accuracy to respond to rapidly changing neologisms and hateful expressions, alongside broader technical and policy efforts across its comment service.

Moving beyond the initial level of automatically detecting profanity and vulgar language based on malicious keywords, Naver expanded the criteria for judging malicious comments to "sentence context" in 2020, actively detecting insulting expressions even in the absence of profanity. Naver has continued upgrading the system to block expressions that cause sexual discomfort, as well as hatred, disparagement, discriminatory expressions, and circumvented malicious comments using symbols and characters.

In addition, the company is actively cooperating with external bodies to establish a healthy communication environment, including reflecting the Korea Internet Self-Governance Organization (KISO) guidelines on hateful expressions in 2023.

"Naver is continuously strengthening Cleanbot's performance to detect not only profanity and vulgar language but also newly emerging expressions of hatred, disparagement and discrimination," Kim Soo-hyang, a leader at Naver, said. "We plan to advance Cleanbot's performance by listening to diverse opinions, including on intensive blocking of content that promotes disregard for life, mocks victims and bereaved families, and spreads hatred."

Original reporting by Lee Jin-seok for Seoul Economic Daily.

AI-translated from Korean. Quotes from foreign sources are based on Korean-language reports and may not reflect exact original wording.

AI KEY

Preview
Korean Corporate Intelligence HubKOSPI · KOSDAQ · 12 sectors

A live, cap-weighted view of every KOSPI and KOSDAQ sector, with same-day Korean reporting distilled by company — built for foreign investors, correspondents and analysts who need to scan Korea before the next session.

Korea Chaebol Tree

Preview
Families Behind the GroupsKFTC May 2026 · DART filings

An English-first interactive map of Samsung, SK, Hyundai, LG and Lotte — built for foreign investors, correspondents and analysts. Korea translates companies into English. We translate the families behind them.