Rakuten, a Japanese e-commerce platform, revealed its intentions to introduce a proprietary AI language model in an interview with CNBC that went live on Monday. It arrives as the behemoth transitioning from fintech to e-commerce seeks to collaborate with other digital companies working on the lightning-fast technology. Before making the AI model available to other companies, the corporation intends to employ it internally to boost operational efficiency.
According to Hiroshi “Mickey” Mikitani, the head of the company, they are developing their own LLM or large language model. Artificial intelligence applications, like OpenAI’s ChatGPT, rely on these enormous algorithms trained on massive data sets.
Since Rakuten is involved in many industries, including banking, e-commerce, and telecommunications, it possesses a wealth of “very unique” data that Mikitani claims it uses to train its LLM. “Nobody has a dataset like ours,” he further stated. According to Mikitani, the business intends to maximize efficiency in operations and marketing efforts by 20% by utilizing the AI model internally.
Similar to how companies like Amazon or Microsoft make their models available to third parties, he plans to do the same for this model. “So we can easily teach them [businesses], package it, and provide the platform for them to completely use it for their business,” exclaimed Mikitani. As an executive, you should know that Rakuten will “have something within a couple of months.”
A representative from Rakuten later stated that the company still needs to set a launch date but may announce its massive language model in the coming months. So far, big IT companies in China and the United States have been releasing their massive language models. Among the most prominent U.S. companies are OpenAI, Amazon, and Google. Baidu, Alibaba, and Tencent are Chinese tech companies that have also released models. Japanese businesses have slipped compared to their American and Chinese competitors. However, their goal is to make up ground as soon as possible.
This month, NTT, a telecommunications company, stated that March will be the launch month for its proprietary LLM. SoftBank’s telecoms unit announced the generative AI computing platform operational in November.
Japanese companies may gain an advantage over their American and Chinese competitors by developing LLMs that are unique to the Japanese language. According to Mikitani, the AI effort will lead to “huge profitable growth” for Rakuten. A Rakuten representative has stated that the company still needs to set a date for when their proprietary AI model will be available for use.
