Clean sources of electricity grew fast enough in 2025 to satisfy all demand growth¹ on the global power grid, and then some—meaning they drove down the use of fossil fuels to generate electricity, albeit slightly. Power emissions fell in both China and India.² Nuclear power made a contribution, with several plants coming online. But the biggest contributor to clean electricity growth by far came from solar panels, whose electricity output jumped by 30 percent last year,³ capping more than a decade of explosive growth. This year, both solar panels and wind turbines are expected to pass nuclear reactors as electricity sources, a historic milestone for the energy transition.

The growth of solar power is so rapid in places like California, Australia and Spain that some grids are flooded with cheap electricity in the middle of the day. More and more, the solution is to build huge, grid-scale batteries that can be charged with solar power at midday, then discharged in the early evening to supply electricity to homes. The cost of batteries has plunged, and their global deployment rose 46 percent in 2025.⁴ We are fast approaching the point on leading grids where batteries charged from the sun will be able to supply more than half the power during the evening peak.

This level of electricity storage was unimaginable a decade ago, but batteries are now becoming a leading technology of the energy transition. This trend, already robust, may be about to accelerate as new technology moves into the marketplace. Grid-scale batteries to date have been based on compounds of lithium, a critical element subject to wild price fluctuations. However, the world’s largest battery company, in China, is coming to market with batteries based on sodium, one of the most plentiful elements on Earth. (The oceans contain quadrillions of metric tonnes of sodium dissolved as salt, sodium chloride.) Lithium-containing batteries can occasionally catch fire, and a handful of disasters have led to some public opposition to the installation of large batteries. Sodium batteries may be safer, and in the long run they are likely to be cheaper in uses like grid-scale storage.

For decades, it appeared as though the greatest bottleneck of the transition would be the supply of clean power: simply building enough wind farms, solar plants and nuclear power stations to make a difference. In 2026, that has become yesterday’s problem across much of the world. It is now becoming clear that the real bottleneck is going to be power grids. They are no longer fit for purpose, and they are not being upgraded fast enough to keep up with the rapid changes going on at both ends of the power lines.

In economic terms, power grids are simultaneously experiencing both a supply shock and a demand shock. The supply shock is coming from the rapid increase of wind farms, solar farms and other sources of renewable energy. A measure of the magnitude of the supply shock is that in many countries, companies seeking to build new renewable capacity face years-long waits merely to obtain permission to connect to the power grid.

The rapid renewables expansion is unquestionably good news, but it is creating a situation that demands careful management. As conventional generators like coal-burning power plants shut down, new equipment may need to be procured to help keep the grid stable. Equipment settings must be reviewed and revised periodically.

The demand shock is also good news, mostly. The fundamental strategy for cleaning up greenhouse emissions in the economy is to electrify as many uses of energy as possible, and then run them on clean power. This is starting to work in many countries, with electric cars displacing demand for oil, and heat pumps in buildings displacing demand for gas. In addition, the rapid construction of data centres to power artificial intelligence models is also placing demands on the grid in some countries, most notably the US. AI programs promise to help cut emissions in the long run, but in the short run, they are a new load that has to be served.

Global power demand was up 3 percent last year, and in the US, where half the new data centres are being built, it was up by 2.8 percent.⁵ This is slower growth than forecast only a couple of years ago, but it is still rapid by historical standards. The rising power demand for data centres poses risks. The companies building them are trying to buy clean power where possible, but they have also proposed a huge new fleet of gas-fired power plants. If allowed, these may lock in high emissions for decades. The biggest data companies are snapping up water supplies; they are so desperate to build they are even stuffing data centres into tents. Their demands on the grid could drive up power prices for consumers if not managed carefully. The situation has already provoked an enormous political backlash in the US, with voters of all political stripes rallying to oppose new data centres. The debate seems to be a proxy for how people feel about AI, with environmental concerns — including the water use of data centres — also playing a major role. AI promises to be a significant political issue in the American elections this autumn.

As the power grid evolves, experts agree that it has to become smarter, more dynamic and more resilient, all at once. The grid needs to get better at moving power in both directions, not just from power plants into homes, offices and factories. The power grids of the future will need to manage fluctuating supply from renewable sources by sending signals to loads — including data centres — that can turn themselves up or down. Imagine a car smart enough to charge itself only when demand on the grid is at its lowest; that sort of thing may be the wave of the future.

These changes are already under way on nearly every electrical grid, but they are not happening fast enough. Not only are the grids physically outdated, but they are encumbered by rules written for another era, run by power companies stuck in the past, and governed by bureaucracies incapable of moving quickly. Tackling this set of problems with smart grid reforms needs to be an urgent priority for every government in the world.

The cost of moving too slowly on grid upgrades has become clear in a handful of major power blackouts. The most recent big one occurred in Spain and Portugal on 28 April 2025. At the time, the Spanish grid was getting almost 55 percent of its power from solar panels.⁶

On that basis and no other, the enemies of the energy transition rushed to blame the blackout on renewable power.

In early 2026, we finally got the truth. The official report on the blackout identified a slew of interrelated failures, with improper settings for both conventional and renewable generators prominent on the list.⁷ The high level of renewable power at the time of the blackout did increase the Spanish grid’s vulnerability to collapse, for a simple reason: a power grid dominated by spinning conventional generators often has enough inertia to allow it to ride through electrical disturbances, whereas a grid running mainly on solar panels might not, unless certain precautions have been taken. The real cause of the blackout was the failure of the Spanish grid operator, Red Eléctrica de España SA, to take those precautions. Special machines should have been added, and settings altered, in ways that would have stabilised the grid at the critical moment. A belated frenzy is now under way in Spain to do that work.

The Spanish blackout is a warning for other countries. The Iberian Peninsula is one of the regions that has proven that power grids can operate with very high levels of renewable energy. But as in Spain, the grids in many countries are antiquated and are not keeping up with the demands that are being placed on them.

Three years ago, at a big climate conference in Dubai, the nations of the world pledged to triple their installed base of renewable electricity by 2030. Judging by the plans and timetables they have since published, they are not on track to achieve this goal, with those schedules implying something closer to a doubling by 2030. Yet the installation of solar power is moving so fast that quite a few countries appear likely to exceed their own targets, and the goal of tripling renewable energy remains within reach.

In fact, the greatest aspiration of the clean-energy movement seems increasingly plausible. That is to create a situation in which developing countries are able to build their own economies on clean electricity, bypassing the fossil-fuel-intensive phase of development and thus restraining the growth of global emissions. The great testing ground of this ‘leapfrog’ idea is likely to be India, whose growth for many years was powered by coal. While politicians there have, in the past, belligerently rejected the idea that India should limit its emissions growth, the facts on the ground are beginning to make that rhetoric look pointless. Solar in India is consistently growing at more than 40 percent a year, and grew 54 percent in 2025. The country is building a single solar farm that will be the largest on Earth, able to supply enough electricity to the Indian grid that it could theoretically power the entire country of Austria. This year, solar will likely supply 10 percent of India’s electricity, and wind about another 5 percent, with both clean sources expected to grow rapidly.⁸ (A subsidiary of Generation, Just Climate, has invested in a major Indian renewable developer.)

The dazzling clean-power growth of 2025 came before the Iran war started on 28 February 2026. That war has, of course, jolted global energy supplies and refocused many minds on alternatives to fossil fuels. Monthly statistics can be misleading and we will not get a clear picture until the end of this year, but preliminary numbers suggest rapidly rising interest in renewable energy across the developing world.

China is the world’s dominant producer of solar technology, and its exports to Africa surged in March, rising by 207 percent in Kenya, 391 percent in Ethiopia and 519 percent in Nigeria compared with February.⁹ These countries are starting from a low installed base, but it will not take many years of such rapid growth for solar to become an important part of their power mix. It has already happened in Pakistan: the disruptions in gas supply due to the Ukraine war in 2022 led to blackouts on that country’s power grid and set off a grassroots push to install solar panels. Solar now accounts for more than a quarter of Pakistan’s electricity, and possibly higher — the change has happened so quickly that analysts have not been able to track it well. ¹⁰

Given the scale of the global solar boom, one might imagine that Chinese producers of solar panels would be wallowing in profits. But China, in its eagerness to dominate industries of the future like solar power and electric cars, has overinvested in production capacity. Moreover, a business in which the price of the product is falling rapidly is a tough business in which to operate — new factories are obsolete almost as soon as they open. Most Chinese solar producers are losing money, and a huge shakeout in that industry seems inevitable, with many companies likely to be forced out of business. It is possible this will lead to temporary price rises in solar panels, as has happened in the past, but the long-term decline seems set to continue.

Chinese dominance in the production of the new technologies has, of course, led to rising global trade tensions. Countries like the US and India are attempting to establish their own production lines for solar modules, though they are still highly dependent on imports of Chinese components. How these tensions will play out in future years remains unclear.

In the US, the administration of Donald Trump took office pledging to slow or halt the clean-energy transition. Mr Trump managed to push through a broad repeal of most of the clean-energy policies that the previous administration, under Joe Biden, had put in place. In situations where the federal government directly controls what gets built, the Trump administration is having success. For instance, the government is cancelling leases for offshore wind farms in areas of the sea floor that it governs, and paying developers billions of dollars to walk away from their projects. These efforts appear certain to set back offshore wind development in the US by at least a decade.

However, the Trump administration’s broader attempts to stop the clean-energy transition are yielding fewer results. On land, developers are still putting up wind and solar farms. The administration is attempting to throw up bureaucratic hurdles to completing projects, but it keeps losing in court. In 2025, with Mr Trump in office nearly the entire year, more than 60 percent of the growth in power demand in the US was met by solar alone, with all clean sources supplying well over 70 percent of the growth.¹¹