The CEO of Japanese fintech-to-e-commerce giant Rakuten has revealed plans to launch an in-house artificial intelligence (AI) language model early next year.
While speaking to CNBC on Monday, Hiroshi “Mickey” Mikitani said Rakuten is working on a bespoke large language model (LLM) that will be trained on a vast amount of “very unique” data exclusive to the company.
A large language model is a deep learning algorithm trained using massive datasets, which enables them to recognize, translate, predict, or generate text and other natural language processing (NLP) content.
Japanese E-Commerce Giant Rakuten Reveals Plans To Launch Business AI Model
Rakuten’s business empire ranges from banking to e-commerce and telecommunications to cryptocurrency trading. Unlike its competitors in the tech industry, the corporate behemoth has access to a diverse dataset due to its expansive ventures.
Mikitani said Rakuten would train its clients on how to use the AI model, “package it” and provide the platform to use it for their business needs.
He also added that Rakuten is on schedule to roll out its proprietary AI model within the next couple of months.
However, a Rakuten spokesperson later clarified that there is no timeline for a launch but there might be an announcement related to the LLM potentially within the timeframe the CEO mentioned.
OpenAI Set To Power Rakuten’s AI Model For Business
Last month, the e-commerce giant selected OpenAI, the developer of the popular AI chatbot ChatGPT, as a strategic partner for delivering “state-of-the-art” AI experiences to its consumers and businesses in Japan and worldwide.
The two companies also unveiled the Rakuten AI for Business, an all-new AI platform provided by Rakuten and powered by OpenAI that is said to empower business customers with “cutting-edge” AI solutions.
Rakuten AI for Business will offer support for a wide range of essential business functions, including marketing, sales, customer support, operations, strategy planning, and engineering.
In a press release, Rakuten explained the different use cases for its business AI.
For example, the Rakuten AI Analyst answers business-related questions with organized data, charts, and insights. The Rakuten AI Agent assists staff in real-time to provide “better, faster, and smarter” customer service. Meanwhile, the Rakuten AI Librarian can synthesize and summarize all of the client’s documentation to answer customer queries in a matter of seconds.
Rakuten AI for Business is currently available through invitation only, with plans to expand the offering next year onwards.
Mikitani expects the AI push to give Rakuten “hugely profitable growth”.
Japanese Tech Firms are Catching up to their American and Chinese Counterparts
While US and Chinese tech giants like Google, Amazon, Microsoft, OpenAI, Baidu, Alibaba, and Tencent have successfully launched their LLMs, Japanese firms are still chasing the crumbs.
Earlier this month, NTT, the telecommunications arm of Japanese multinational holding company SoftBank, announced that its proprietary large language model will be available to the public in March 2024.
Meanwhile, SoftBank began operations of its generative AI computing platform last month. According to reports, chip-making giant Nvidia is set to power its AI computing platform.
Back in August, the fintech launched SB Institutions, a new company that is dedicated to researching and developing “homegrown” LLMs specialized in the Japanese language.