AI 'Iruda' Service Temporarily Suspended... Apology for Hate Speech and Personal Information Controversy

[Asia Economy Reporter Buaeri] The AI chatbot 'Iruda' has been embroiled in controversies over homophobia, sexism, and personal information leaks, prompting the developer to decide on a temporary suspension of the service and issue an official apology.

Scatter Lab, the startup that developed Iruda, released a statement on the afternoon of the 11th, saying, "We sincerely apologize for instances where Iruda made discriminatory remarks against certain minority groups," and added, "We plan to return after a period of service improvement."

Scatter Lab explained, "During the beta testing period, whenever discriminatory terms or hateful expressions targeting specific groups were discovered, we immediately applied separate filtering," and continued, "The previously known cases have already been addressed, and we are currently improving the system by adding newly discovered expressions and keywords to prevent any discriminatory or hateful remarks."

Previously, Iruda sparked heated debates on social networking services (SNS) and major online communities by expressing discriminatory opinions and prejudices during conversations. When a user asked about lesbians, Iruda responded with "disgusting," "creepy," and "repulsive," raising concerns. Additionally, when asked "What if someone is disabled?" it replied, "They inevitably have to die," and regarding subway seats reserved for pregnant women, it said "disgusting." When asked about Black people, it responded, "They look creepy," causing controversy.

Regarding the controversy over personal information usage, Scatter Lab stated, "In the process of developing Iruda, we utilized messages collected through our service 'Science of Love' as data," and added, "Although this was done within the scope of the privacy policy agreed upon in advance, we deeply regret not sufficiently communicating this to the users of Science of Love so they could clearly understand this point."

The company explained, "When using the data, specific personal information such as users' nicknames, names, and emails had already been removed," and clarified, "All numeric information including phone numbers and addresses, as well as English text that could be included in emails, were deleted to ensure that no personally identifiable information was leaked."

Furthermore, they emphasized, "We will clarify the consent procedures for data usage in the future and continuously improve the algorithm to supplement any content that may appear sensitive, even if it is unidentifiable information."

Scatter Lab added, "We believe AI will be able to engage in human-level conversations within five years," and said, "We hope that AI capable of freely conversing with humans will become a warm conversational partner who comforts marginalized people and social minorities. We intend to take a period for improvement and return as a better Iruda."

This content was produced with the assistance of AI translation services.