Japan Bets $6B on a Sovereign AI Model and 10 Million Robots
2 min readJapan is making one of its boldest bets yet on artificial intelligence. The country’s Ministry of Economy, Trade and Industry, known as METI, has committed up to 1 trillion yen, roughly 6.2 billion dollars, over five years to Noetra, a national consortium built to develop homegrown AI models and put millions of AI-powered robots to work. It is a clear push for technological independence in a field dominated by American and Chinese firms.
Inside the Noetra Consortium
Noetra is anchored by four of Japan’s biggest names: SoftBank, Sony, NEC, and Honda. Roughly 44 companies are expected to take part, with an initial tranche of 387.3 billion yen flowing in the 2026 fiscal year. Working alongside the national research institute AIST, the group plans to release its first foundation model this fiscal year, then improve it annually using data drawn from Japanese manufacturers.
Betting on Physical AI
What sets Japan’s plan apart is its focus on “physical AI,” the software that lets robots sense and act in the real world rather than just chat. The government is targeting 10 million AI-equipped robots deployed across 18 sectors, including restaurants, food manufacturing, and medical care, by 2040. The aim is to offset a shrinking workforce and aging population with machines that can help fill labor gaps.
The strategy leans on Japan’s traditional strengths. Sony brings sensors and imaging, Honda brings robotics and mobility, and SoftBank brings capital and a long history of AI investment. Rather than compete head-on with chatbots from OpenAI or Google, Japan is aiming at a niche where its hardware expertise gives it an edge.
Why It Matters
Sovereign AI has become a national priority across Asia, and Japan’s scale of commitment stands out. A homegrown foundation model reduces reliance on foreign providers for sensitive data and critical infrastructure. If the robot rollout succeeds, it could become a template for other aging economies searching for ways to stay productive. The first foundation model, due this fiscal year, will be the initial test of whether the money translates into results.