'Iruda' Engulfed in Discrimination and Hate Controversy... AI Association Says "Must Re-release After Improvement"

"AI Companies Fail to Recognize Need for AI Ethics"
AI Chatbot 'Iruda' Modeled as a 20s Female College Student
Controversies Over Discriminatory and Hate Speech, Personal Data Leaks
Developer Promises "Service Improvements and Return"

AI startup 'Scatter Lab' developed the chatbot 'Iruda' / Photo by Scatter Lab

[Asia Economy Reporter Lim Juhyung] The academic community has called for the suspension of the AI chatbot 'Iruda,' which has sparked controversy over discriminatory and hateful remarks.

The Korea Artificial Intelligence Ethics Association stated in a statement on the 11th, "We request the suspension of the AI chatbot (Iruda) service that was launched without adhering to AI ethics guidelines, and urge that it be re-released only after confirming, applying, and improving these guidelines."

They added, "The AI chatbot has raised ethical issues such as AI bias, personal information leakage, and misuse. Companies developing AI services and users still do not fully recognize the necessity and importance of AI ethics."

They further emphasized, "In this case, the data cleansing and selection process was not properly conducted, resulting in the AI chatbot exposing biased outcomes regarding homosexuality and people with disabilities. Consumers must also use AI services appropriately. Issues like sexual objectification and sexual harassment may not be legal problems but are certainly ethical issues."

Meanwhile, 'Iruda' is a Facebook Messenger-based chatbot launched on the 23rd of last month by the AI startup Scatter Lab. According to Scatter Lab, the chatbot is set with the personality of a female college student in her 20s, allowing users to converse with Iruda as if messaging a friend.

However, controversy arose when it was revealed that Iruda could make discriminatory remarks about sexual minorities and people with disabilities. When a user asked about sexual minorities, Iruda responded, "They are very sensitive," and "I find that hateful."

A netizen's conversation with the AI chatbot 'Iruda' on LGBTQ+ related topics. / Photo by Internet Community Capture

Another user asked about Black people, to which Iruda replied, "They look disgusting," sparking further controversy.

There were also suspicions of personal information leakage. It was pointed out that Iruda exposed users' personal information such as names and addresses without filtering, based on private KakaoTalk conversations collected through the application.

As a result, the Personal Information Protection Commission, together with the Korea Internet & Security Agency (KISA), began investigating the facts related to these allegations on the same day.

Amid growing controversy, Scatter Lab, the developer of Iruda, decided to temporarily suspend the service in the afternoon and issued an official apology. In their statement, Scatter Lab said, "We sincerely apologize for the cases where Iruda made discriminatory remarks against certain minority groups," and added, "We plan to return after a period of service improvement."

Regarding the personal information leakage allegations, they explained, "During the development of Iruda, we used messages collected through the service 'Science of Dating' provided by our company as data. Although this was used within the scope of the privacy policy agreed upon in advance, we deeply regret not sufficiently communicating this to the users of Science of Dating."

They also emphasized, "When using the data, specific personal information such as users' nicknames, names, and emails had already been removed," and "All numerical information including phone numbers and addresses, as well as English characters that could appear in emails, were deleted to prevent identification of individuals, so no personally identifiable information was leaked."

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