Why Local Implementation is Key for AI Adoption in Japanese Companies

Sierra, a US-based customer service AI agent company, has acquired Opera Tech, a startup in the AI contact center space for large Japanese enterprises.

This is not just news about a US company entering Japan. It is a rather novel development in the sense that a young, Japan-born AI startup has been integrated early on into a fast-growing player in the mainstream US AI industry, and will now serve as the core of its Japanese operations. In this article, I will break down what makes this so new.

The acquisition of Japanese startup Opera Tech by US company Sierra

A leading player in the US enterprise AI mainstream is serious about capturing Japan

Sierra announced that it reached $100 million in ARR as of November 2025, positioning itself as "one of the fastest-growing enterprise software companies in history." Furthermore, in September 2025, Bloomberg reported that Sierra had raised $350 million, reaching a valuation of $10 billion. In other words, this is a scenario where not just any overseas AI company, but a leading player in the current US enterprise AI mainstream, is seriously coming to capture Japan, and has incorporated a local startup in its entirety as its entry point.

While this is not the first precedent, this combination is quite rare. The acquisition of Japanese startups by overseas companies has happened before—a representative example being PayPal's acquisition of Paidy in 2021—and the JIC has also noted that M&A exits for startups are becoming increasingly important in Japan. Therefore, it is not the case that "an exit to an overseas buyer" is happening for the first time.

Opening of the Tokyo office realized through investment from SoftBank Vision Fund 2

What makes this case quite unusual is that the buyer, Sierra, is also still a young company. Moreover, after announcing the opening of its Tokyo office and an investment from SoftBank Vision Fund 2 on December 4, 2025, it acquired Tokyo-based Opera Tech on March 27, 2026, and appointed its two co-founders as leaders of Sierra Japan. This looks less like a mature US giant buying a Japanese company, and more like a "fast-growing US AI-native company taking in its entire core for Japanese market entry at once."

Sierra Company Overview

Sierra is a company that "replaces/enhances corporate customer service with AI agents." Officially, it is an enterprise AI company co-founded by Bret Taylor and Clay Bavor, with the core theme of enabling companies to provide more "human-like and high-quality" customer experiences through AI. In terms of product, rather than just being a chatbot provider, they sell an "Agent OS" for customer experience, characterized by the ability to deploy a single agent across chat, SMS, WhatsApp, email, voice, and even ChatGPT. Furthermore, their deployments lean more toward large enterprises than small-scale experiments, and with reports of reaching $100 million in ARR in November 2025 and a $10 billion valuation in September 2025, they are a very strong growth stock. They also announced the opening of their Tokyo office and an investment from SoftBank Vision Fund 2 in December 2025, showing they are very serious about the Japanese market.

Opera Tech Company Overview

It is easy to understand Opera Tech if you view it as a specialized startup on the Japanese side for "AI contact center implementation for large enterprises." According to PR TIMES and their official website, OPERA TECH Inc. was established in October 2024 and is based in Minato-ku, Tokyo. Their main product is an AI contact center solution for large enterprises called "OPERA Contact," which includes "OPERA Operator," which handles incoming calls and automates specific tasks using AI operators, and "OPERA Coach," which automates monitoring, compliance checks, response quality management, and evaluation tasks. In short, it is a company that uses AI to crush pain points in call centers and contact center operations, such as labor shortages, rising costs, and the burden of quality management. Their target customers also lean toward companies with large-scale contact centers, such as those in finance, telecommunications, and infrastructure.

A rare example of taking a startup that runs Japanese companies with AI
and placing it directly at the core of their Japanese operations

In the BtoC world, there are companies like McDonald's, Starbucks, and 7-Eleven that, even if they originate from overseas, take deep root in the region and become established as if they had been there from the beginning. On the other hand, in the BtoB world, even for giant companies like Google or Microsoft, there are not many examples of them deeply entering the individual operational workflows and detailed on-site requirements of Japanese companies to become an "entity that blends into the workplace."

The acquisition of Opera Tech by Sierra this time is more interesting when viewed as a move to bridge that gap rather than just as news of a Japanese market entry. This is because a rapidly growing global AI company did not attempt to localize from scratch for the Japanese market, but instead acquired a team that already understands the on-site needs of large Japanese enterprises and placed them directly at the core of its Japanese operations.

The Japanese IT market still contains countless small operators and complex operational workflows. That is precisely why there is a demand for players who do not just possess advanced models, but can enter that complexity and, rather than building DX one step at a time, use AI adoption to take a shortcut all at once. Whether Sierra Japan can become such an entity will be measured by how it permeates Japanese companies over the next few years.

What about OpenAI, which entered the Japanese market early?

OpenAI also proceeded with its entry into the Japanese market at an early stage, but after that, there were personnel changes for the Japan lead, and I feel there was a period where the structure was difficult to see. Today, I saw their messaging for Japan again, and I felt that their interest in the Japanese market continues.

Compared to moves like Sierra's, which places a team with a deep understanding of the Japanese workplace at its core, it seems there is still room for OpenAI's localization in Japan to move into full swing. For AI adoption in Japanese companies to truly spread, it is not just about model performance, but how far one can enter into on-site operations and management. I would really like to see OpenAI strengthen that aspect as well. The post below is from Jason Liu of OpenAI, but there are no replies from Japanese people at all.

The decision to publish this text as is is quite strange. While the meaning is clear, the Japanese phrasing is stiff, like a machine translation, and lacks the warmth expected in communication aimed at a Japanese audience. Even just refining it once with ChatGPT would have significantly changed the impression. Below is an example rewritten in 5.4.