Japan warmed up to AI late but that could turn out to be an advantage as this technology evolves

Diffusion, meanwhile, is finally picking up. During the first quarter this year, AI adoption in Japan rose 3.4 percentage points, over three times faster than the global average, according to Microsoft’s AI Economy Institute Q1 report. Separate data from Docomo found usage nearly doubled between this February and last.

In a country long admired for hardware but less so for software, engineers are using AI to narrow the gap. The Microsoft report found that developers in Japan uploaded 129% more code changes to GitHub than a year earlier, compared with the global average of 78%—a signal that the technology is already changing how coders work.

Global tech companies have noticed. A new wave of entrepreneurship is taking place in Japan and “we want to be part of the story,” Dmitry Shevelenko, chief business officer at Perplexity AI, told me recently. He’s far from alone as the likes of OpenAI, Anthropic and others are opening offices.

Yet the country remains relatively insulated, and there is still hesitation about going all-in with foreign institutions. Companies that want to be part of Japan’s AI transformation should act like long-term stakeholders, hiring local staff and offering serious partnerships.

This could also mean fostering deeper collaborations with universities and research communities. Scientists remain one of the few groups in Japan that still command high public trust, according to an Edelman survey. AI firms that want credibility should borrow less from Silicon Valley’s hype machine and more from Japan’s institutional ecosystem.

Still, as Shevelenko puts it: “If you’re successful here, it’s a real signal to the rest of the world that you’ve built something of incredible quality.” He’s correct that this discernment is an edge.

Business leaders should also focus less on promises of future abundance and more on urgent, practical problems (one reason Silicon Valley’s pitch no longer resonates so easily in the US).

In Japan, tourism is an obvious example. AI could do much more than translate menus. It can build smarter itineraries, redirect travellers to regions that still need visitors, smooth congestion and help spread the record inbound demand beyond the same crowded destinations.

Tokyo has favoured a light-touch regulatory approach, with the goal of becoming the most AI-friendly nation in the world. But policymakers should also recognize the advantage of playing catch-up. Japan can watch, learn and adapt what works. This is especially true in higher-risk bets like Agentic AI, where you hand off more autonomy to machines.

None of this guarantees success. Japanese corporate culture remains cautious and the current consumer curiosity does not automatically translate into broader transformation. Business leaders and policymakers are right to encourage startups, but they must extend support to smaller and medium-sized firms. In addition, mid-career and older workers are still less likely to use the technology though they remain a large swathe of the labour force.

Japan’s most famous tech investor, Softbank founder Masayoshi Son, is making one of the world’s boldest bets on the technology with a $60 billion commitment to OpenAI. Whether that proves visionary or reckless remains to be seen, but the time horizon is instructive. On Softbank’s most recent earnings call, its CFO said the company is looking at the next 30 years, adding, “The AI revolution has only just begun.”

For Japan, the long view is the right one. The current frenzy creates a FOMO pressure to spend big now and justify later. But if AI is truly a transformative general-purpose technology—more akin to the steam engine that propelled the Industrial Revolution than a new smartphone app—the winners will not be the ones who sprint first. They will be those who do the marathon task of embedding it deepest across the economy. ©Bloomberg

The author is a Bloomberg Opinion columnist covering Asia tech.