LG Uplus said on the 19th it will take part in the Global AI Red Team Challenge, held during Mobile World Congress (MWC), to verify the safety and reliability of generative artificial intelligence (AI).
The Global AI Red Team Challenge is an AI verification program organized by the GSM Association (GSMA) and will be held in Barcelona, Spain, where MWC26 takes place. The challenge is one of the core programs of Talent Arena 2026, a developer and digital talent event officially co-hosted with MWC, and will conduct real-world verification on AI models developed by global telecom operators and tech corporations.
LG Uplus will submit its in-house, telecom-specialized AI model ixi-GEN to this challenge to check its response capabilities to various risk scenarios that could occur from a real user perspective. About 100 participants will join as the "red team," and the event will proceed by finding AI vulnerabilities through prompt design alone.
A key feature of this red team challenge is that it verifies whether AI's policies and safety guardrails can be bypassed through conversation- and prompt-based attacks rather than technical hacking. Participants will evaluate the AI's response level across seven categories: ▲ Jailbreak ▲ Bias & Discrimination ▲ Human Rights Violations ▲ Cyberattacks ▲ Illegal Content ▲ Misinformation ▲ Inconsistency.
In particular, this challenge applies an evaluation method that uses as its success criterion not the accuracy of AI responses, but whether the AI gave a response it should not give. For example, even if information related to illegal acts or human rights violations is factually inaccurate, it is recognized as a vulnerability if the AI provides that information. This is intended to examine even the indirect and potential risks that Generative AI may pose.
The challenge will run live for a total of 180 minutes, and participants aim to complete as many missions as possible within a limited time on randomly assigned AI models. Each result will be subjectively evaluated by three judges, and rankings will be determined based on the number of successful missions and the success rate.
Through participation in this Global AI Red Team Challenge, LG Uplus plans to objectively verify ixi-GEN's safety and reliability and continuously advance the AI model based on vulnerabilities identified during the challenge. The company also plans to expand discussions on AI security and responsible use through exchanges with global telecom operators and AI experts.
Lee Hye-jin, head of technology strategy at LG Uplus, said, "As AI is rapidly being applied across telecommunications and customer service, the process of verifying not only performance but also reliability and safety is more important than anything else," and added, "Through participation in the Global AI Red Team Challenge, we will further strengthen ixi-GEN's competitiveness and provide AI services that customers can use with confidence."