Singtel, Deutsche Telekom, e&, SK Telecom, and SoftBank, the founding parties of the Global Telco AI Alliance (GTAA), have signed a joint venture agreement at TM Forum’s DTW24-Ignite.
The GTAA, at the Mobile World Congress (MWC) Barcelona 2024, announced its aim to establish a JV. The GTAA aims to make investment to codevelope and launch multilingual Large Language Models (Telco LLM) specifically tailored to the needs of telecom companies.
The JV will have equal investments from the founding parties to support its working capital requirements to develop multilingual Telco LLM that to help telcos improve their customer interactions via digital assists and other innovative AI solutions. The GTAA did not disclose financial details.
The JV will look at deploying innovative AI applications tailored to the needs of the founding parties in their respective markets, enabling them to reach a global customer base of approximately 1.3 billion across 50 countries.
The Telco LLM will be multilingual comprising Korean, English, German, Arabic, and Bahasa, among other languages. The launch of the JV is subject to customary regulatory approvals.
The second Global Telco AI Roundtable (GTAR) was held at DTW24, reaffirming the GTAA’s dedication to driving innovation and collaboration in the telecoms industry. The GTAR highlighted the key progress achieved by the GTAA, as well as the AI governance systems and AI use cases reshaping the telecom landscape.
SKT’s CEO Ryu Young-sang highlighted that the GTAA is well-positioned to be an active player in the global AI ecosystem as businesses are keen to reshape their business strategies through AI. “New business opportunities will be created and enhanced levels of customer experiences will be achieved,” Ryu Young-sang said.
The founding parties showcased their applications of an LLM for telco, focusing on contact centre and infrastructure use cases. They demonstrated how an AI-enabled LLM can enhance contact centre operations by generating real-time references for agents during calls and automatically handle post-call tasks.
They also illustrated the model’s ability to comprehend domain-specific documents, data, and processes to provide answers to infrastructure operators’ questions, streamlining their workflows.
