I’m going to be honest with you. I worry a lot.

I worry about all sorts of things. Money, friends, family, if my voice is weird as you’re listening to it... but there’s one thing I’ve been worried about since I was a teenager in the early 2000s. The climate.

<Newsclip: “The world’s going to hell in a handbasket.”>

<Newsclip: “It is a wakeup call.”>

<Newsclip: “It is the greatest con job ever perpetrated on the world in my opinion.”>

<Newsclip: “So how do we leave this pathway to disaster?”>

Those anxieties have only gotten worse. In recent years, it has felt like climate action has stalled. In 2024, the world passed the 1.5˚C ‘threshold’ set in Paris for the first time. Shortly thereafter, Donald Trump returned to the White House and on the first day of his Presidency set about reversing climate policies of the previous administration, vowing to promote fossil fuels. In Europe, right-wing politicians have taken aim at climate goals, with the EU weakening previously ambitious environmental policy. All while it’s getting hotter and hotter. In May there were record breaking temperatures in my home of the UK , then in June those records were broken again, twice in the same week in fact, with temperatures in some places across the continent more than ten degrees hotter than they were between 1991-2020. That was the story all over Europe. Without climate change, it’s extremely unlikely Europeans would have had to endure these heatwaves. And globally, scientists of all stripes have expressed doubts that any government has done enough to avoid the worst impacts of climate change.

So, it feels like we’re failing on climate, heck, I can even look outside my window and see evidence for that. But what’s weird is that even as climate change is affecting me more, I find it harder to engage with.

As things get worse, I’m having trouble staying optimistic that we, you me and everyone, can do anything about it. And that’s an odd position to be in, as a science journalist. Climate change is one of the biggest stories of our time.

Every week, we pick out stories to cover on the podcast. And I always want to report the latest climate science, and I do, but I find it hard. Sometimes it just feels like more bad news at a time when our audience, and I, could do with less. I feel hopeless.

But is that actually right? Am I justified in that feeling? As the one thing I know about science is often the truth is a bit more complicated than we think. So, while the narrative can feel overwhelming negative, is that really what the data says?

To try and shake myself out of this climate funk, talking to some of the scientists and campaigners on the front lines of climate research, and what I found was a story of: Tipping points…

That has the potential to move this energy transition over the edge.

I don't think we've had this level of uncertainty about a technological shift before.

I think there's this positive feedback loop by which all of our individual level actions can lead to bigger collective actions.

…and snowballing effects.

The system is changing at exponential speed.

This is Cooked? a three-part miniseries for Nature, where I try and find out whether we are well… cooked. Is it game over for climate or are there reasons for hope? I’m your host Nick Petrić Howe. And in this first episode, I decided to start with the data, and I had a good idea of who might be able to help me with that.

As a data scientist, I'm biased. And I would say, look at the data to understand for these big problems, where they're coming from, why they're happening, and importantly, what we can do to solve them.

In the past year or so, I’ve found myself struggling with anxiety about the climate. But a couple of years ago a book came out that seemed to fly against what I had assumed was the case, as it stated, backed up with oodles of data, that it isn’t the end of the world. So, I knew I wanted to speak to its author — Hannah Ritchie.

What I wanted to do in that book was, think, twofold.

That’s Hannah, the author of ‘Not the end of the world’ and a data scientist at our World in Data and the University of Oxford.

One was that I was trying to write it to my former self, who was very, very pessimistic, wasn't sure if there was really a future there for me and present where we are now. And the point is not that these problems are solved in any way, they're still huge problems, but we have started to make progress and there's lots of opportunity for us to make more progress. My point is not that we should be complacent and optimistic that these problems will solve themselves. I think it's to kind of maybe get rid of a bit of the cynicism that there's nothing that we can do and actually no one is doing anything because I don't think that's true. And I think the second kind of motivation for doing that is to look at the data and understand if we want to solve these problems, what are the really key and big things that we need to do to solve them?

The book looked at seven huge challenges that humanity faces, including climate change, examining each through the lens of data, to understand the progress we have and haven’t made on them, and what we should do. So I was curious to hear from Hannah about how we were actually doing on climate change.

If you go back 10 years, the world we were sitting in when the world signed the Paris Agreement was that the alternatives to fossil fuels were much more expensive than fossil fuels. For solar and wind were way more expensive. Batteries were extremely expensive. Electric vehicles were basically not on the radar at that point. So it seemed very, very hard to me at the time to see how the world would actually progress on this. What we've seen over the last decade is this huge drop in the costs of these low-carbon energy sources. And that now means that they're cost competitive with fossil fuels, right? So you have the kind of short-term incentive of picking the low-cost solution, now adding up and aligning with the long-term issue we're trying to solve, which is climate change.

These days renewable energy is very cheap. Let’s take solar energy as an example. Between 2009 and 2025, estimates from the investment firm Lazard suggest solar energy prices have dropped to less than a sixth of what it used to cost on average. This number even includes the continued costs of running solar plants, making them, along with wind, the cheapest form of energy.

This change has helped Hannah go from pessimistic to optimistic about the future of climate change. It’s not just that the costs have tumbled, more and more this ‘clean’ energy is actually being implemented as well.

We now also see that many countries are now rolling out clean energy at an increasingly fast pace. So it's not like it's just that the cost has dropped. You can actually now start to see that translating into action.

Hannah also thinks that this electrification of the world has other benefits too.

What's really key about that electrification lever, and I think is under appreciated, is how much you gain there in terms of efficiency. You know, it's not just we want to do this because it saves us carbon emissions, which it will if we were powering that through low carbon power, but also you get huge efficiency gains from this. like driving in an electric car uses about a third of the energy of driving in a petrol car, right? So you actually also get rid of lots of this energy waste that we currently have in the system.

Worldwide, adoption rates of electric vehicles are increasing. Around a quarter of every new vehicle sold is electric, and in some countries this rate is much higher. In Norway 97% of new cars sold were electric, while in China 53% were. Now, many people are critical of electrical vehicles for various reasons. Mining of lithium for use in batteries can be very environmentally damaging, for instance. And using public transport is better than having a personal electric vehicle most of the time. But when it comes to their climate credentials, electric cars beat conventional every time.

The world has also been built in many places to favour vehicles and in some areas public transport is not a viable option. So if we want the most people to use the least climate emission causing form of transport, electric vehicles will be part of the mix.

But it’s not just data on clean energy and electrification that gives Hannah hope. She also points to progress that’s been made in making steel and concrete production more climate conscious. They’re maybe a bit less obvious in our day to day lives, as most of us don’t work on big construction projects, but these core components of buildings come with big emissions. Cement production, for example, accounted for around 6% of global carbon emissions as of 2024 according to the International Energy Agency.

But this is down from a peak in 2021, due to slowed growth in construction, and is projected to decrease further as improvements are made to the recipe for cement, or alternatives to it entirely are created.

These solutions haven't been scaled yet. They're at a very, very small level, but I think we still have made progress there and there are actually solutions starting to come online and starting to get cost competitive, but it will take some time for them to build up to a kind of level where you would actually start to see dents in the data.

As Hannah said, the point of her book and this examination of the data is not to allow us all to pat ourselves on the back and say ‘job’s a good ‘un’. There’s still a lot that we need to do, and in many ways some of these changes aren’t happening fast enough. Rather the point is to not make us feel despondent and hopeless, as that helps nobody.

And there are actually reasons for optimism. Falling prices in renewables, increased electrification and more climate-conscious construction are all starting to make dents in the data according to Hannah.

So maybe I don't need to be so worried after all. assuming of course that we can keep this up.

Renewable energy growth has continued in the US, but all it's doing is making the rise in emissions less bad.

There's a new disruptive and energy-intensive technology out there that could stall climate progress. Artificial intelligence.

The level of debate about this technology is massive. It's actually really fascinating because I don't think we've had this level of uncertainty about a technological shift before.

Whenever I open social media these days, I see a lot of scepticism about artificial intelligence. I don’t have to scroll far to come across a video of someone pretending to be a large language model powered chatbot who says silly things while chugging litres of water. But the backlash against the technology is spilling out of social media and pouring into real world legislation. We've seen the first moratoriums on the building of data centres, the facilities that house the infrastructure for AI and other internet services. And often the cited reasons are environmental — water and electricity use.

So that got me wondering, if AI is so energy intensive could it stall climate progress? Especially as many of these data centres are using fossil fuels to meet their energy needs. For the sake of this podcast, I’m just going to focus on energy demands, as that’s what’s more relevant to the climate, and put other environmental concerns to one side for now.

To do so, I wanted to put those memes aside as well and find out what the data actually says. So, I reached out to an expert on data centres and their energy use to go through some of the numbers.

My name is Ketan Joshi. I'm an independent researcher and analyst, working on energy and data centres and fossil fuel phase-out.

Ketan has been working as energy analyst for many years, but more recently has been focusing on how much energy data centres consume, particularly for artificial intelligence. I wanted to know how a growth in data centres would affect the climate, but he explained to me that that’s a quite a difficult thing to work out.

We have thousands of projections of data center growth into the near future. But the level of variation that you see in these projections makes me almost want to never use them. and the reason behind that is really because some projections basically take a linear pathway, you know, it'll grow to maybe four to five percent by twenty thirty. I've seen ones that are up to like twenty percent of global energy consumption for data centers, you know, in the coming decades. And I find that to be a little bit silly. And the reason behind that is really because some projections basically take a linear pathway. You know, it'll grow to maybe 4 to 5% by 2030. I've seen ones that are up to like 20% of global energy consumption for data centers, you know, in the coming decades. And I find that to be a little bit silly. And the reason behind this wild skew of future projections is because there is an inherent assumption about the future of artificial intelligence and the shape and form that it takes in the near future. The way that we tend to formulate projections just falls apart with something like this where there is an inherent uncertainty as to what the technology is, what it's going to be used for, whether it's hype, whether it's real, whether we go for high energy consumption models or low energy consumption models. There's like 10 crisscrossing vectors of uncertainty that make it really, really hard to say what's going to happen.

The future is, unfortunately, unpredictable. It’s hard to know what AI will look like in 10 years, or even 10 days for that matter. So, it’s tough to fit into our projections of the future and how it may impact the climate. As a result, Ketan likes to focus on the here and now to see what we can understand about the technology. To start, it’s important to note that while AI gets a lot of attention, it currently forms a small part of energy usage. The International Energy Agency, for instance, estimates that data centres use — for everything, not just AI — around 2% of the world’s energy. As Ketan says, that may grow — perhaps by a lot, perhaps by not so much. It’s part of the uncertainty with the technology.

According to Ketan’s analyses, though, AI and by that, I mean the generative kind that powers chatbots, may be particularly energy intensive.

Something that I've found in the past few weeks actually, having looked at some of the new newer corporate statements. So for instance, Microsoft analyzed their latest statement, I analyzed Google and Amazon. And all three of those companies are using more energy to earn the same amount of revenue, if that makes sense.

Rather than working towards more efficiency as computing has done in the past, Ketan thinks that companies are using more and more energy to try and make cleverer chatbots, or autonomous AI agents. Google and Microsoft for example, have acknowledged that they’ve increased their energy usage as they build out AI capacity.

One thing to note here is that different use cases of AI have different energy demands. Techniques like Machine Learning which often gets labelled AI, don't use as much energy as generative AI like Large Language Models. And then within generative AI, things like a written response from a chatbot needs a lot less energy than generating an image or a video.

But the concern for the climate is that all this energy needs to come from somewhere. And if there isn’t sufficient renewable energy in place, it will come from fossil fuels.

Figuring out how much data centres may be using fossil fuels, though, is not easy. Ketan explained to me that when companies report their power consumption it’s not quite as simple as them just stating what proportion of renewable energy versus fossil fuels they’ve used. Similarly to carbon offsets, companies can purchase certificates from renewable energy providers and then use them to show that their emissions are lower than they are. Microsoft recently stopped using this method, which has been widely criticised, and in their most recent report there was a large rise in their emissions.

The companies can also have positive effects on renewable energy. With their vast resources they can fund new wind and solar farms. Microsoft, for example, has announced a $6 billion agreement to build AI infrastructure powered by renewables in Norway. Deals such as these can also then be included in companies’ reports on what kinds of energy they use.

So, untangling all of this is pretty tricky. With that in mind, Ketan explained how energy use has changed in the US, where there’s the most data centres in the world.

There's a new report from the energy information administration, updating on 2025 data. And of course, you know, emissions have gone up. They haven't gone up crazy. Like they haven't gone, you know, it's not like it’s skyrocketing, you know, right angle, but they have gone up and renewable energy growth has continued in the US, but all it's doing is just making the rise in emissions less bad.

The report Ketan mentioned says that US energy-related CO₂ emissions went up by about 2% in 2025. That may not sound like a lot, but it’s against a backdrop of emissions falling, so any rise is a cause for concern for the climate. And like Ketan says, renewable energy adoption has continued apace, so the fact that there were more renewables, but emissions still went up implies that more energy was coming from carbon-emitting fossil fuels.

This is backed up by a report from the Energy Institute, a UK-based professional body that monitors energy use. The report said that US emissions had increased in 2025 due to higher coal use, and that electricity demand had risen faster than supply driven in part by data centres and artificial intelligence. An independent research firm, the Rhodium group, also identified a rise in carbon emissions in the US and cited a particularly cool winter, more coal use and increased demand in electricity from data centres as reasons.

Of course, data centres could be powered by renewable energy, but building this supply takes time and if we have a sudden increase in energy demand, due to AI for instance, then fossil fuels could be used to plug the gap. Google, for example, has acknowledged that it’s expanding its AI development on a grid that is undersupplied with clean energy, but still remains a large funder of clean energy projects. Google still plans to reduce their emissions, but right now, they’re using more fossil fuel energy.

Ketan would like to see more transparency from the tech companies creating AI on its environmental impacts, so it can be scrutinised more easily, along with more regulation from governments. He also thinks that everyone has a role to play deciding on what terms we use this technology.

It's basically and this this like spectrum of power, you know, ranging from very little power, which is like I'm just one person who makes decisions about, you know, energy use, to institutional power, which is pretty substantial, like, you know, an institution saying, I'm not going to subscribe to your chatbot until you can actually be transparent about where the energy comes from, how it's used, et cetera. and then to the most powerful, which is like you know, an energy company saying, we will not supply you with energy unless you actually make it make sense, and you have like public support and societal support for the actions of your company.

Ketan is actually optimistic about the future, as it seems that the general public are beginning to understand that they do have the power to demand better from these companies. Environmental, cost and climate concerns are being raised against AI, and politicians are listening — you can look to the growth of moratoriums on data centres to see that. To be clear this isn’t just because of climate, opponents of data centres describe them as being ugly, straining local resources and not providing many tangible benefits to their residents. These are issues that affect people no matter what side of the political divide they sit on. And this bipartisan opposition has meant that politicians have started taking action.

But that touches on another thing I wondered about. What are governments and nations actually doing? Are they meeting the goals they’ve set out for climate? And it’s certainly felt to me like there’s been some backsliding recently — with right-wing politicians taking aim at climate goals.

So, what does the data say about how the nations of the world are doing?

Well, it's better than we used to think 10 years ago, but it's definitely not there where it should be.

If you’ve ever engaged with climate change, you’ll know that it’s a global issue. The whole world is getting hotter and that affects all of us. Because of that, nations often get together and come up with various pledges on what they’re going to do about that. What comes out of these pledges is ultimately what we, as a global society, are doing about climate change.

And I knew a scientist who keeps close tabs on this, so I gave him a call.

Hello. Niklas Höhne, New Climate Institute, calling in from Cologne, Germany.

I met Niklas a few years ago at COP26, the enormous climate conference run by the UN. He co-founded the New Climate Institute, a non-profit that works on climate policy, where he oversees analyses on climate policy. I was there reporting for Nature and I asked him, even back then, if things were hopeless. I remember that he didn’t coddle me or make me despair. He gave a nuanced picture. So, I caught up with him to see where we are now.

So while 10 years ago we thought global greenhouse gas emissions would rise forever and would lead to a temperature increase of three and half degrees by the end of the century, we now think that with currently implemented policies, they will stabilize and decline and then lead to around two and a half degrees by the end of the century. So a full degree better than we thought only 10 years ago. So that's remarkable. That's really good. But it's not enough. The Paris agreement says well below two or in the direction of 1.5 and there we are far away from it. And also, we have to say that in the past years, global greenhouse gas emissions have still increased. So despite all of the efforts, is still more fossil fuels being burned each year and therefore we have more greenhouse gas emissions each year and that means not only are we not stopping climate change, we are fuelling climate change. So, we are accelerating the speed of change each year by adding more greenhouse gases to the atmosphere than we had in the year before.

The main driver of this drop in the projected temperature increase

is something that Hannah already touched on — the speed of renewable energy production.

Prices are falling and Niklas thinks this may push us to precipice to start some exponential change.

Niklas Höhne: I think the last 10 years we've been surprised each year about the speed of renewable energy, electrification and the transformation that has been started. So, I assume that also next year and the years after we will be positively surprised because the end of that development of the technologies is not there. Yeah. So, prices still decline. New battery technologies are invented. Renewables are incredibly cheap compared to 10 years and are still declining. And what we as humans often underestimate is exponential change. The system is changing at exponential speed. And that means each year more solar is implemented than we thought before or more batteries are implemented, more electric cars. And most of the models, they are still kind of assuming a linear change and underestimate the speed of the transformation. And that's why I think I'm quite optimistic about our future projections that this renewable energy transformation will actually win against the fossil fuels because it's simply exponential change. And that always has to win over a linear trend, which is the one described by the fossil fuels.

You can see a similar exponential change if you look at the history of cars.

At the turn of the twentieth century there was blight on cities — horses. According to the London Transport museum in 1900 there were 300,000 horses in London. But just 14 years later they’d all but disappeared. This sudden departure of the city’s workhorses helped humanity solve a pretty relevant problem for a climate podcast — pollution

Horses produced a lot of… waste — you know the kind of waste I'm talking about — that was very literally clogging up streets. The solution, ironically, given the topic of the podcast, was cars. Motorized vehicles had actually been around for a few years in the early 20^th Century, but it wasn’t really until 1908 when the Model-T ford came out, the first car affordable to a large number of people, that they really took off. In 1910 there were 8000 cars registered in London, in 1920 there were 190,000, the next year there were 250 000 And as cars took off, streets once clogged with horse manure instead became roads for motorcars.

That may seem like a pretty inevitable change, but if you were a Londoner in 1900, you’d probably wouldn’t have believed that the world could be so different in a few decades.

Perhaps renewable energy will be the same for us. It may seem like we’ll never overcome our addiction to fossil fuels, but as prices tumble, maybe we’re standing on the cliff edge of that exponential change.

So there are certainly positives for the climate. The speed of the energy transition has taken many experts by surprise. But beyond that, there’s still a long to go. And when it comes to how countries are actually sticking to their climate pledges, it was a bit of a mixed bag.

We regularly look at who is basically the best, which is the best country around the world. And the places one, two, three are always empty. So, no country is doing everything correct. But in each area, you can see quite positive things. We have electric mobility, which is shooting off and was shooting off in Norway because Norway supported it significantly with a lot of money. But now it's basically the better way of doing things. So even countries like Vietnam or Nepal have very high adoption rates for electric cars. They sell more electric cars than conventional vehicles in these countries. And that is remarkable because it shows that this is the most economic way of having individual mobility. Then there's other areas, buildings and industry, which really are a bit behind. And it's very difficult to find really good examples where countries are really doing well. But OK, I think also in buildings and in industry, this idea of electrification is something that will now take off as well. So in buildings, it's heat pumps for heating, which are the new normal. So, there's huge opportunity to do more in those areas as well.

You may have noticed that some of the biggest emitters haven’t been mentioned there. And to be honest, they were some of the ones I was most worried about. China and the US.

China is the world’s top emitter according to the International Energy Agency, but it seems that their emissions are finally starting to peak — meaning that the greenhouse gases they emit may be starting to reduce. According to some analyses, they have already peaked. This is likely because they have implemented a huge amount of renewable energy. Last year, China installed more solar capacity than the rest of the world combined, although this boom in installation is slowing down a bit now.

And as Niklas explained to me, China has also been instrumental in producing what the world needs for renewable energy — they’re the leading manufacturer of solar, wind and battery technology. But he did impress on me that that wouldn’t have been possible without the policies of several countries, especially in Europe, promoting renewables and therefore creating the right market conditions for China.

And then there’s world’s second biggest emitter and the largest historical contributor to climate change. The US. There I was worried about the influence of the most recent administration. Trump’s Team has dealt a severe blow to the world’s ability to tackle climate change, by walking away from Paris Agreement on Climate, for the second time. They also repealed the ‘endangerment finding’ that set out that climate change is a public threat and has been used to enact climate policy. And they have spent billions trying to undo climate projects, giving the companies behind them refunds if they instead invest that money in fossil fuels. These actions, the administration say, would save the United States money by removing excess regulations, lower bills for families and help maintain the US’s leadership. But they are counter to what climate scientists say we should be doing — we need to stop burning fossil fuels and rein in our emissions.

So, I was expecting a pretty negative response from Niklas when I asked about how the US is doing. But he had a few surprises for me.

So for the US as such, it's the second largest emitter and the Trump administration has basically dismantled all climate policies that were there from the federal government in the US. And that means now that the emissions in the US will still continue to decline, but will decline slower.

This first surprise, is that while the actions of the Trump administration will likely increase the emissions that the US will produce, that is against a backdrop of overall decline. Even though there's been a recent uptick in emissions in the past year, the US's emissions have been steadily falling since the early 2000s and that will likely continue. Albeit not as fast as it would if the Trump administration was more pro climate. Ultimately, the economics of clean energy are hard even for the most powerful country in the world to fight against — it’s getting harder and harder to justify the use of fossil fuels when the alternatives are cheaper.

The second surprise was an unexpected way that the current US administration may have unintentionally pushed for a move away from fossil fuels.

Which is the closure of the Strait of Hormuz. And that created a significant energy crisis with very, very high fossil fuel prices for oil and gas. And that created a push towards energy efficiency and renewable energy. And that effect will be significant. It's very difficult to quantify that right now. But from our point of view, it's very clearly that that will has the potential to move this energy transition over the edge, so to say, and over the tipping point that it will be self-propelling because all the countries that currently suffer significantly from fuel shortage is not even a price, a high price is really a shortage that simply have not enough. They will now really, really think about alternatives and have already. So, some countries have already increased their renewable energy targets. China has reduced their oil consumption by 10 to 20%, depending on which source you look at within only 12 months. So, there's significant changes on the way that will reduce global greenhouse gas emissions.

As Niklas mentioned, data on this is hard to get, and this is still an evolving story. Time will tell what the impacts of the war in Iran may be. There have in fact been moves towards and away from renewable energy as a result of surging oil and gas prices — with nations like India and South Korea burning more coal in response to the higher gas prices, while people in the Philippines have taken to installing rooftop solar panels at record rates to avoid the high oil prices. So, a true accounting of the effect of the closure of the strait of Hormuz will take some time. But Niklas’s assessment is backed up by an increased sale of electric vehicles and a record high in solar exports from China, and an analysis by Carbon Brief that suggests that global emissions may fall in 2026. This has led some to wonder whether the war in Iran will be a tipping point for energy — that it will finally end humanity’s addiction to fossil fuels.

Overall, while there are clearly some areas in which the world is behind on its targets, it’s perhaps not as dire as I thought. In terms of renewable energy, adoption is accelerating. True, the current pace won’t be enough to meet our climate goals. We need to stop emitting carbon altogether, and soon if we're to meet the 1.5° target set in Paris, and even then we'll probably have to take some carbon out of the atmosphere. But Niklas, even with all the caveats, was optimistic about the future.

We have all the tools at hand and even now it's economic and the obstacle is not technology, the obstacle is not cost. The obstacle is our inertia in the system and that some lose something and some others gain something. And that is something we as a society really have to overcome.

So, while there’s still a long way to go, there are reasons for optimism. Renewable energy has outperformed many researchers expectations, and continues to do so, in spite of the backlash against it in some quarters. Which, I’ll be honest, does make me feel better. The data tell me that it’s not completely hopeless. When cities were overwhelmed by horses and their waste, cheap motor vehicles changed things in a matter of decades. I’m hopeful that renewable energy can do the same here.

But this is what the data says, largely about what we’ve already done. I was also curious about what we should do in the future. What are some of the big ideas that can help tackle climate change? That’s next time.

We could sit here and say well people should just do this because they should just care about the climate and reducing their emissions. But I think having very good affordable substitutes for people to switch to is just really, really crucial.”

This episode was written and produced by me, Nick Petrić Howe. Dan Fox was my editor. And thanks to Hannah Richie, Ketan Joshi, and Niklas Höhne for speaking to me for this episode.