The 5 worst ideas of the 21st century – and how they went wrong

These are our selections for the worst fumbles of the 21st century: ideas that were great, but got twisted or misused and didn’t deliver on their original promise.

Bitcoin

For years, it was a constant refrain in the tech world: “Put it on the blockchain.” Bitcoin, a cryptocurrency that was invented in 2008 and went mainstream in the late 2010s, brought blockchain technology into the public eye, and it has exploded ever since.

On the face of it, bitcoin seems like a good idea – it is a currency that isn’t overseen by any government or banking regulator, but rather by a public ledger: the blockchain, in which all transactions are recorded. Copies of the ledger are distributed on a network of computers around the world, kept secure by a combination of sophisticated cryptography and the fact that each copy of the ledger can be checked against all the others to ensure its accuracy.

This article is part of our special issue on the 21 best ideas of the 21st century.
Browse the full line-up here

To get bitcoin, a person can either buy it like any other asset or “mine” it. Mining a bitcoin essentially consists of solving a cryptographic puzzle, and the more bitcoins that have already been mined, the harder it becomes to solve the puzzles necessary to acquire the remaining ones. That’s where the problems start to come in.

In the early days, anyone with a basic computer could mine bitcoin, but it now requires enormous computing power, so bitcoin miners tend to set up entire rooms or warehouses full of processors to crunch the numbers.

Obviously, warehouses full of constantly running computers guzzle energy, in a way that was unheard of until artificial intelligence became common. Researchers at the University of Cambridge calculated that in 2023, bitcoin alone used more than 120 terawatt-hours of energy. That’s more than some entire countries, and nearly half a per cent of global energy use for that year, even without taking into account the many other cryptocurrencies that have popped up since bitcoin’s rise.

The main energy suck is bitcoin mining, but making a transaction, even a small one, eats up power, too. “The average carbon footprint of a person taking a flight from London to New York is smaller than a single bitcoin transaction,” says Alex de Vries at the Free University of Amsterdam. “It would be more environmentally friendly to just take a bag of cash and fly it there.”

While it is possible to reduce this energy consumption by changing the architecture behind bitcoin, as its closest competitor Ethereum and some other cryptocurrencies have done, its distributed nature means that this would require consensus across a majority of bitcoin owners. “There’s no other way we could reduce this much energy consumption overnight,” says de Vries. “It should be a no-brainer [to reach that consensus], but apparently it’s not.”

And all that is before even considering the difficulty of actually using bitcoin in day-to-day life. “If you want to use it as a currency, to actually pay for stuff, most of the time you can’t because the capacity of this network to process transactions is limited,” says de Vries. “It’s like a digital collectible, but there are plenty of other things you can collect – in fact, if you look at recent history, you’re better off investing in Pokémon cards than bitcoin.”

Bitcoin is, in effect, a huge betting pool that gobbles up unthinkable amounts of energy and resources and is best used for illicit transactions on the dark web. Better to bet on Pikachu.

Leah Crane

Social media

It used to consist of cat videos, genuine debate and connection, and citizen journalism on the biggest moments of the day, whether they were natural disasters or revolutions. And memes. Lots and lots of memes. Early-stage social media was both fun and functional. But as profit for its tech company creators soared, the user experience decomposed into a fetid yet addictive slop. Aside from asbestos, CFCs and single-use plastics, nothing has made such a pendulous swing from useful to harmful as social media.

Take Instagram. Once upon a time, it served up a chronological feed of images posted by a carefully curated list of people. We could keep up to date with friends, follow the output of artists we enjoyed and share our hobbies. Now, we’re force-fed a never-ending stream of influencers, adverts and lowest-common-denominator content that algorithms know will rile us into grudging engagement.

It is perhaps safe to say that social media was a very good idea indeed, but that unfettered capitalism isn’t – and we allowed the former to be monopolised by the latter.

At its best, social media allowed people with impossibly niche interests or problems to connect, forge friendships and create support networks. It spontaneously spawned protest groups like #MeToo, the Arab Spring and Black Lives Matter, which have shed light on inequality and fostered public debate.

At its worst, it has spread misogyny, hatred and extreme politics, allowed cyberbullying and created unique social problems we never foresaw and are yet to unpick. Ask a high-school teacher whether social media has created or solved more problems, and the answer will be brief and unequivocally negative.

“I don’t think we were ready for it as a society, to have the information overload,” says Andrew Kaung, who has worked at tech giants including Meta, Google and TikTok.

Kaung says social media wasn’t intentionally made awful. Rather, it was done methodically over time, the side effect of a process designed to maximise how long people stayed glued to apps – and therefore how much money could be extracted from them. As is often said, if you’re not paying for a product, you are the product.

Engineers and data analysts found that negative posts reached a larger audience than positive posts, says Kaung. These squeezed more comments out of people and kept them engaged for longer. So it is easy to see how social media slid from reasonably sensible political discourse to polarised, vitriolic tribalism, and how it began to spread hateful content that deliberately enraged users. It is simply more profitable.

Perhaps the only way to solve the problem is to break the link between social media and profit. Remove the levers that allow the people who run these tools to worsen the experience for everyone else and ramp up their own profit, and we might have the beginnings of a path back to the early joy of social media. We would have to cut out the incentive for algorithms to maximise engagement while disregarding whether people are actually enjoying the service or getting practical benefits from it.

“The problem is not so much the technology, which I think still could be salvaged. It’s the business model,” says Graham Murdock at Loughborough University, UK. “If we don’t do something decisive, then I think we’re going to see more and more negative repercussions, and I don’t think they’re going to be controllable.”

What does that decisive action look like? Perhaps taking the power out of the hands of the few and giving it to a wider group. There certainly exist open-source social networks that run on distributed models just as cryptocurrencies do, but they have struggled to gain traction. Legislation to curb the worst instincts and motivations of technology companies has been introduced, but the vast wealth of these firms allows them to lobby the teeth out of it. For me, the solution was simple: delete the apps.

Matthew Sparkes

Carbon offsets

It sounds like such a simple idea. You take a flight that produces a lot of carbon dioxide, but pay to have some trees planted that will take up the same amount. Or a company whose activities produce lots of CO₂ can invest in projects to reduce or avoid other CO₂ emissions – or even to remove it from the atmosphere – so no more ends up in the atmosphere overall. It’s a win-win, right?

In practice, “offsetting” CO₂ emissions in this way gets really complicated, really fast. For starters, the success of offsetting projects must be assessed on the basis of what would have happened otherwise, but what’s assumed as the counterfactual can wildly change the outcome. Let’s imagine you buy some land and decide to reforest it, and to sell so-called carbon credits based on how much carbon the trees will take up from the atmosphere.

The assumption here is that no trees would grow without your actions. But maybe that land would have been reforested naturally, creating a more diverse forest than a planted one. And maybe that natural forest would have been more resistant to droughts and diseases than a planted monoculture forest. In this scenario, your actions could result in there being more CO₂ in the atmosphere, rather than less.

Offsetting projects are also supposed to have additionality – that is, they wouldn’t happen without carbon credits being sold – but this is easy to fake.

“With a good enough accountant, there seem to be unlimited ways of proving additionality,” says Myles Allen at the University of Oxford. For instance, renewable energy projects can be made to look as if they wouldn’t be financially viable without selling carbon credits.

Even where offsets do make a difference, the benefits may be hugely exaggerated. For instance, many offsets are based on emissions avoided by saving forests. But carbon credits may be sold based on an entire forested area even if only 1 per cent of a forest was at risk of being destroyed. And that can have real consequences. People might, say, take more flights because they think their emissions are being offset, when that is far from the case.

Yet another danger of offsetting is that it can be an alternative to investing in greener tech. It is much better to permanently lower your emissions by, say, buying an electric car than to offset a few long journeys to ease your conscience.

With nature-based offsets, even with the best of intentions, there is also no way to guarantee that they will remain offsets for as long as required. Going back to our example, let’s say the trees you plant thrive for a couple of decades, but then a wildfire, drought or disease kills most of them. Or maybe you die and your greedy kids sell the land to a developer. Whatever it is, the end result is that there is no less CO₂ in the atmosphere.

Allen argues that in the long term, the only way to permanently remove CO₂ from the atmosphere is to put it in some form of geological storage. This kind of offsetting is now becoming available, with a few companies trying to remove CO₂ directly from the atmosphere and store it underground. The trouble is that it is far more expensive than other forms of offsetting – and the price may not come down that much.

Supporters of offsets argue that some projects do deliver on their claims and that offsetting can have wider benefits, such as helping preserve biodiversity, as well as limiting emissions. But how can you tell? Allen and his colleagues have set out principles that “good” offsetting schemes should follow – but the first is that reducing emissions is always better than offsetting where possible.

Michael Le Page

Alternative fuels

Fossil fuels cause global warming. So, if we switch to alternative fuels, the problem is solved, right? If only. In fact, the push to switch to alternative fuels has so far been largely a disaster that is causing immense environmental damage and driving up hunger and poverty while failing to reduce emissions – with one massive exception.

In the broadest sense, the term “alternative fuels” refers to any fuel that isn’t derived from oil. There is a huge variety, but the main types are natural gases, hydrogen, biofuels and synthetic fuels.

The idea with synthetic fuels is to use renewable energy to turn CO₂ from the air into, say, jet fuel. It’s great in theory, but in practice it remains difficult and very expensive, so only tiny quantities are being produced. Scaling up production would also require huge amounts of renewable electricity that, for now at least, would be better used directly. It is way more efficient to use electricity to power cars and heat pumps, for example, than to convert electricity into a fuel and then use that.

And while swapping coal- or oil-based products for natural gas can reduce CO₂ emissions, natural gas is still a fossil fuel. If switching to gas delays the complete switch away from fossil fuels – the only way to limit further warming – it can be counterproductive.

Almost all hydrogen is made from natural gas at present, so while burning it produces only water, its production releases lots of carbon dioxide. Even if the production of green hydrogen from renewable energy could be scaled up, for most purposes it doesn’t make sense. Green hydrogen is just another synthetic fuel and, again, it is better to use the electricity directly.

Then we come to biofuels, such as bioethanol made from maize. To understand why they are such a bad idea, think about what has to happen to produce “sustainable” jet fuel from palm oil. It would require growing more palm oil, already a major driver of deforestation and habitat loss. That means clearing more rainforests and losing the rich wildlife they harbour.

In addition to the carbon in their trees, rainforests often have thick layers of peat that degrade when the land is drained. All this stored carbon ends up in the atmosphere. Then there’s the fact that farming is one of the biggest sources of greenhouse emissions. The end result is more CO₂ in the atmosphere, not less.

More demand for palm oil also means rising prices – the growing use of biofuels is one of the factors driving food inflation – which means more people struggling to afford food.

“If just a quarter of global aviation fuel by 2050 came from crop-based biofuels, massive amounts of additional farmland would be needed,” says Richard Waite at the World Resources Institute in Washington DC. “It would lead to large-scale deforestation and higher food prices, and would be a disaster for our efforts to fight climate change and prevent biodiversity loss.”

Biofuels made from waste products such as pig fat can be genuinely sustainable, but there is simply not enough unused waste for large-scale production. For instance, more than 80 per cent of the used cooking oil turned into biodiesel in the European Union is imported, and some of it has turned out to be fresh oil fraudulently sold as used.

But let’s end on a positive note. It might not be a liquid or a gas, but in some countries, electricity is officially an alternative fuel, and it really can be clean and green. The best way to slash emissions is to make electricity greener and cheaper, and electrify everything possible.

Michael Le Page

Effective altruism

Imagine that I could whisper in your ear a simple principle that would make you a better person. Follow it and you can help save multiple lives every week, simply by giving away the cost of a takeaway coffee.

This is the starting point of an idea called effective altruism (EA), which says we should rationalise our charitable giving by working out the most impactful ways of saving lives and supporting those causes. You are probably wondering what could be wrong with that – and the idea’s seemingly obvious appeal is precisely why it has been so damaging.

This concept has roots that twist back a long way, notably to philosopher Peter Singer, and in particular his 2009 book The Life You Can Save. There, Singer argued that if you saw a stranger drowning in a pond, you would wade in and save them – it would cost you next to nothing. By the same token, he argued, people from rich Western nations can and should help people living in poverty by donating small sums to fund effective interventions, such as vaccines or anti-mosquito bed nets that help stop the spread of malaria.

The idea really crystallised in the public consciousness around 2012. It was then that philosophers Toby Ord and William MacAskill at the University of Oxford founded the Centre for Effective Altruism. The phrase stuck and the idea started to get traction, attracting vast sums from, among others, the billionaire couple Dustin Moskovitz and Carrie Tuna, who were interested in giving away the fortune Moskovitz had made as a founder of Facebook as effectively as possible. Organisations such as GiveWell also began to evaluate charities and assess which had the most positive impact per donated dollar.

So, what’s the problem? Few would argue EA has never done any good at all, but there are big concerns nonetheless. For example, one spawn of the EA movement is the idea of “earning to give”, meaning would-be altruists are sometimes better off taking a high-paying job than trying to directly address, say, poverty. The thinking is that these high earners can then give away much of their wealth to good causes.

That might sound good in principle, but it doesn’t always happen. The earning-to-give mindset became popular in Silicon Valley in the 2010s, and one of those who espoused it was Sam Bankman-Fried. Following the collapse of Bankman-Fried’s cryptocurrency exchange FTX in 2022, the company tried to claw some of these charitable donations back.

And that gets us to perhaps the most problematic thing about EA. Moral philosopher Alice Crary, who has written extensively about EA, says the crux of her problem with the idea is that the concept of doing “the most good” needs an awful lot of unpacking – it isn’t something that can be boiled down to an efficiency calculation.

The idea that we can maximise the good we do based solely on scientific evidence simplifies morality in a way that can seem deeply attractive, and it is yet profoundly misleading. “The world that we engage in as moral beings needs to be understood, for instance, historically and culturally,” says Crary. “It can’t be understood through a purely abstract lens of the sort that allows for quantification.”

She points out that it is hard to put numbers on the impact of some obviously good things. Take the example of training doctors, she says. It is a far less cost-effective thing to do than simply give out vaccines. Yet if we don’t have doctors to build good relationships with poor communities, we can end up in a situation where one child’s bad reaction to a vaccine turns a whole community against life-saving treatments.

The idea of maximising the good we do can also take us to some bizarre and manifestly undesirable places. For example, one prominent idea is that the future existential risk to humanity from a superintelligent AI is so serious that research that might help avoid this is at least as important as work to curtail poverty today. Many Silicon Valley insiders bought into this idea and funnelled money to AI start-ups such as OpenAI and Anthropic. So, although earning to give may provide the sheen of being related to some higher moral purpose, we aren’t talking here about the antimalarial bed nets that Singer envisaged.

There are also plenty of stories of effective altruists who believe so strongly in their obligation to maximise the good they do that they end up neglecting simple things like doing the dishes or caring for family. Perhaps without realising it, these people have chosen a world in which philanthropic efficiency is prioritised more than close, loving relationships with other humans. “They have envisioned a world in which relationships are drained of meaning,” says Crary.

So do all the good you can, by all means – just don’t believe anyone who tells you there is an easy principle that can tell you how.

Joshua Howgego