
Noetra, a government-backed company, is getting more than 380 billion yen ($2.33 billion) in government funding for its first year as Japan pushes to lock down key technologies through a foundational model for physical artificial intelligence and robots.
Who Holds the Levers
Hironobu Tamba, Noetra’s CEO, called the company the "last chance" for Japan in an interview. That’s a blunt admission from inside the machine. The state is bankrolling the project, 44 companies are lining up behind it, and the whole setup is meant to give Japanese companies access to a "reliable and safe foundational model" where they can manage data themselves. Tamba said, "We've been able to bring together people who have been working on AI in Japan," and added, "Honestly, I think this may be our last chance." He is also an executive at SoftBank.
Noetra is procuring 27,500 Nvidia Rubin chips and plans to begin construction on AI infrastructure in April 2027, with operations starting in June of the following year. The company’s backers include SoftBank, Honda and Sony. That’s not some scrappy neighborhood workshop. It’s a state-backed industrial bloc, stitched together with public money and corporate muscle, aiming to shape who gets access to the tools and who gets left waiting at the gate.
Japan’s government is also backing projects including foundry venture Rapidus, which plans to make cutting-edge chips. Prime Minister Sanae Takaichi's administration wants to deploy 10 million AI-enabled robots by 2040 in manufacturing, shipbuilding and nursing. The language is all about self-sufficiency and access to cutting-edge technology, but the power sits where it always does: at the top, with ministries, executives and industrial planners deciding the future for everyone else.
Who Pays for the Machine
The people at the bottom don’t get to vote on whether 380 billion yen goes into this project. They get the consequences. Japan, which has leading makers of industrial robots, is trying to improve domestic supply of key technologies while China and the U.S. remain locked in an AI race. That race is framed as national survival, but the costs and risks are socialized while the control stays concentrated.
Noetra’s pitch is that Japanese companies will be able to manage their own data through its model. That sounds tidy enough until you remember who owns the infrastructure, who gets the chips, and who decides what counts as "safe." The company says it will provide access to a foundational model, but the arrangement itself is built on public funding, corporate backing and a supply chain that runs through Nvidia hardware.
The government’s 10 million robot target by 2040 reaches into manufacturing, shipbuilding and nursing. Those are not abstract sectors. They’re workplaces and care settings where ordinary people will live with the decisions made in boardrooms and cabinet offices. The apparatus calls it progress. The rest of society is expected to absorb it.
Taiwan’s Separate Script
Taiwan’s top tech official, Wu Cheng-wen, said the island’s drive to build its robotics industry will strengthen its AI technology supply chain and ease the strains of an aging society. He drew a distinction from China’s military-oriented approach. That’s the official line: one state says it’s building for society, another for the military, and both still route power through institutions that ordinary people don’t control.
An industrial humanoid robot was displayed at Foxconn’s annual tech day in Taipei in November 2025. The display was a neat symbol of the whole arrangement: corporate spectacle, state strategy and technological ambition presented as inevitability. The people expected to live with the results get the least say in how the system is built.
Japan’s robotics push and Taiwan’s industry drive are being sold as answers to demographic strain, supply-chain weakness and geopolitical competition. But the facts on the ground are simpler. Governments are pouring public money into machine power, corporations are crowding in, and the future is being assembled from the top down.