Rakuten's 1.8B users get AI that reads minds, not just clicks

Rakuten isn’t just dabbling in AI. The Tokyo-headquartered giant, with its sprawling ecosystem of 70+ services and 1.8 billion members, has gone all-in on a partnership with OpenAI to fundamentally rewire how it extracts value from its most critical corporate asset: data. Yusuke Kaji, General Manager of AI for Business at Rakuten Group, put it bluntly: “For us, the data asset is the key corporate asset.” This isn’t about a flashy chatbot for consumers. It’s a deep, multi-pronged integration using OpenAI’s API, including GPT-3.5, Code Interpreter, and retrieval-augmented generation (RAG), to tackle core business problems.

The results are already tangible. A customer service process that once forced users to wait days for a ticket resolution is now automated, using RAG on an internal knowledge base to deliver instant help. For shoppers, Rakuten is building a feature that synthesizes hundreds of user reviews into structured, digestible summaries of key topics. The B2B side is seeing an even more profound shift, where consultants are now armed with AI to provide merchants and enterprises with actionable insights from market analyses and sales trends, transforming a previously manual, slow process of data crunching.

But the story here is as much about trust as it is about technology. In Japan, Kaji warns, a misstep with a large-scale AI deployment could have a “chilling effect on innovation for many years to come.” That cultural and regulatory reality drove an extensive security assessment and a thorough due diligence process before Rakuten chose OpenAI as its partner. The commitment to privacy and security is the non-negotiable foundation for every project, with Kaji insisting on “guardrails in place to the highest standard.”

Looking ahead, Rakuten is eyeing OpenAI’s real-time voice and vision capabilities to turn internal meeting audio into action items and translations, and to build more conversational AI experiences. This reveals a strategic flywheel: conversational AI doesn’t just answer questions; it captures a rich, direct signal of user intent that makes mere clicks and impressions look like a pale proxy. Kaji sees this new data stream as the key to understanding a user’s actual pain point, creating a loop where better data leads to better service. For a company that views AI as an existential tool to empower small businesses and create an “AI empowerment company,” the mission is clear: capture that direct preference data, or risk being left behind.