In July 2025, Anno Takahiro, the leader of the new political part Team Mirai, won a seat in the House of Councillors (Upper House) election in Japan. The party has been described as “technocratic” and focuses on subjects such as e-government and the use of AI. Anno himself is an AI engineer and science fiction author who graduated from Tokyo University.

I was curious about the party’s policy on energy. Looking at their manifesto, it is clear that their emphasis is on ensuring a steady supply for energy hungry AI data centers but not much thought seems to be given to the environment.

A massive amount of electricity is essential for the full-scale adoption of AI. It is estimated that the world will need an additional amount of electricity equivalent to the entire country of Japan in the short period leading up to 2030. For Japan to master AI and use technology as a driving force for growth, securing a stable and large amount of electricity is a prerequisite.

However, Japan is poor in fossil fuel resources, and relying on thermal power generation would result in a large-scale outflow of national wealth due to fuel imports. Furthermore, the limited plains of the country pose geographical constraints on the large-scale introduction of renewable energy.

Therefore, it is necessary to accelerate technological development and capital investment that will enable Japan to maximize the use of all energy resources available in the country while simultaneously securing large-capacity power sources and realizing a zero-emission society.
(“Policy Manifesto 2026: Energy”)

Their main objection to fossil fuel-based power generation is not the disastrous effects on the climate but the outflow of money for fuel imports.

A goal of “zero emissions” is not mentioned until the very end of the paragraph, almost as an afterthought.

Outside the power sector, transport, heating and industrial processes all currently use fossil fuels and therefore produce carbon emissions. Electrification is the only way to decarbonize all of these, but none seem to merit a mention in the manifesto, unlike AI. To me, that’s a lopsided approach.

Renewables are mentioned primarily to claim that their scope in Japan was limited. In reality, wind turbines do not need limited flat land and they coexist quite well with agricultural use. They work especially well on the coast lines, of which Japan has plenty. Akita, Aomori and Hokkaido in particular have huge potential for both on-shore and off-shore wind. Rooftop solar is far from reaching a saturation point in Japan. Sitting on the “ring of fire”, Japan has some of the highest geothermal potential in the world.

Team Mirai view of renewables is stuck in the past (Mirai means “future” in Japanese) when they claim:

A significant increase in renewable energy will pose major challenges in terms of both the burden on the public and practicality.

The public burden due to renewable energy surcharges is expected to reach 2.7 trillion yen in fiscal 2024 and 3.1 trillion yen in fiscal 2025. The introduction of additional renewable energy will lead to an increase in the public burden.

Japan does have a system of renewable energy surcharges on electricity bills that was used to get solar projects online when equipment costs were still much higher than they are today. Back then, renewable energy suppliers were guaranteed a feed in tariff high enough to cover those higher capital costs. The renewable surcharges are to pay back the investments from years ago. They do not reflect the current competitive situation. Nowadays solar and wind are the cheapest sources of electricity in most of the world, which is why the International Energy Agency (IEA) estimates that:

Over 2025-2030, renewables are expected to meet over 90% of global electricity demand growth.
(Renewables 2025, IEA)

Pakistan doubled its solar capacity from 2024 to 2025 not because of any feed-in tariffs or renewable energy subsidies, but simply because photovoltaic panels are so cheap, their power is cheaper than grid power from hydro or coal.

When Team Mirai warns of the cost of an “unreasonable expansion of renewable energy”, it misrepresents the actual cost basis of renewables vs. fossil and nuclear power. The problem for Japan is not that solar and wind would consume huge amounts of subsidies — they don’t. It’s that the Japanese grid will need significant upgrades to move power from regions with the best potential for renewable energy to the regions with the highest demand. While China has been building thousands of km of HVDC lines to bring hydroelectric, solar and wind power from rural areas to the cities and industries, Japan still has very limited power exchange capacity between eastern and western Japan, which use different mains frequencies, or between its major islands like Honshu and Hokkaido.

So what is Team Mirai’s recipe for expanding power generation? Like the LDP and several other mainstream parties, it wants to restart as many nuclear power stations as quickly as possible. Beyond that it wants to bet on unconventional nuclear technologies such as nuclear fusion and small modular reactors (SMRs):

We will strengthen investment in research and development of nuclear fusion technology , demonstrate our international superiority in nuclear fusion technology, and prepare for a fundamental solution to long-term energy problems.
We will support the technological development and dissemination of next-generation nuclear power (SMR, high-temperature gas-cooled reactors, etc.) with an eye toward the late 2030s and beyond.

Japan is one of the major participants in the international fusion project ITER. However, under correct timescales (which already moved by many years from plans drawn up before the project was launched), actual fusion experiments that may produce energy (deuterium-tritium fusion) are not expected before the year 2039. Nuclear fusion is a highly complex technology that will still take decades to work at scale, if it can be made to work at all. All this complexity has a cost. Meanwhile the cost of renewables and storage has been falling year after year. Whatever fusion technology wins the race, it will have to compete with energy from a free fusion reactor in the sky whose output can be captured by anyone using ever cheaper photovoltaic panels. It is far from clear if nuclear fusion will ever become as cheap as renewables.

Electric vehicles will add huge amounts of storage that can be used for vehicle to grid (V2G), which addresses a lot of the intermittency of renewables. A typical EV battery stores several days’ worth of electricity consumption of an average home. With an expanded HVDC grid, local intermittency is of much less concern when different regions can share output at any one moment.

SMRs are becoming fashionable at the moment, but so far they are not in commercial use anywhere. It remains to be seen if they can actually realize the cost savings their proponents promise. We have to remember that historically, nuclear reactors started out with much smaller sizes and then every generation grew in output per unit. Nuclear submarines and aircraft carriers essentially are powered by SMRs, but these designs are not cost effective for commercial, non-military use. It will be around 2030 that the first new SMRs are expected to come online, with larger scale deployment not achieved before the middle of the 2030s. It’s only when they are mass produced that they have a chance to produce power more cheaply than conventional reactor designs.

To summarize, Team Mirai’s energy policy is not driven by scientific or economic realities. It grossly underestimates the potential of renewable energy in Japan while betting on technologies with an uncertain economic outlook that may simply come too late and at too high a cost to solve the problem of decarbonizing the Japanese economy before the country is devastated by climate change.