A general view of the Bitcoin booth at the 2015 International CES at the Las Vegas Convention Center on January 8, 2015 in Las Vegas, Nevada. Photo: Getty Images
Show Hide image

Slowly but surely, Bitcoin appears to be falling apart

Once seen as a better investment than gold, the digital cryptocurrency is experiencing some severe existential threats.

Been paying much attention to Bitcoin lately?

For many people (well, most people, let's be honest) the digital cryptocurrency hasn't been a particularly newsworthy topic for at least a year - at least, not since the bubble which saw the price of one bitcoin peak at $1,242 in November 2013, before popping, rising a bit again in early 2014, and then declining again. Regardless, a number of online retailers now accept Bitcoin instead of fiat currency from customers, and there are even major sporting events sponsored by Bitcoin companies.

The media (and I include myself here) became bored of Bitcoin as it became just another tech novelty, like Google Glass or virtual reality. Not necessarily ignored, though - major crises, like fraud in the large Bitcoin exchanges, or a black market being shut down, are still interesting - but post-bubble there's been very little of relevance to those outside of the fields directly related to it, like online retail, or cryptography.

This has been just fine for many of Bitcoin's most passionate advocates, who saw the media's scrutiny as a primary factor in causing its earlier speculative bubbles. One of the defining quirks of Bitcoin is that it's simultaneously a currency and a commodity (to the chagrin of tax authorities worldwide), though, and this caused some tension among those who hope to see it become a serious threat to, say, Western Union: the long-term health of Bitcoin is generally seen as dependent upon it becoming seen as being as boring and reliable as cash, but at the same time speculators driving up the price were by far the best way to both raise awareness among the lay population and convince others to invest in "mining" it.

But, of course, that speculation has given Bitcoin a price volatility that has scared away exactly the kind of people who need to be confident their currency isn't going to devalue overnight, without warning, and without the kinds of consumer protection that central banks and nation states offer. That's been the problem from the start, really. Bitcoin isn't going to replace any fiat currencies as long as it feels intuitively safer for most people to keep their savings in a bank account instead of in a digital wallet.

This brings us to the current crisis in Bitcoin: far from widespread adoption giving it resilience and reliability, the system may be starting to fall apart. The price of a Bitcoin versus the dollar has been falling steadily since that brief post-bubble "recovery" in early 2014. Here's a chart from bitcointicker.co of BTC versus the US dollar going back roughly 18 months:

The trend is clearly down, and has been for a while, the occasional rally excepted. Worse still for those holding BTC, too, it seems to be speeding up - particularly in the last week:

The price of a bitcoin is skittering around just above $150 as of writing, but there's no reason to think it won't keep falling, for two reasons.

Firstly, as with many previous crashes, there's been a spike in the volume of trading on Bitcoin exchanges - particularly this week. But unlike the situation with previous crashes, this is taking place in the context of that year-long decline in BTC price. There are a lot of people who bought big in Bitcoin back when it was worth more than gold, and they've spent most of the last nine months hurting for it; most infamously, Bitcoin was one of the few investments in 2014 to give a worse return than the rouble.

There's plenty of anecdotal evidence to suggest that groups (like hedge funds, for example) which bought huge numbers of bitcoins are now dumping them onto the market as they try to get out of what looks to be a lost cause. Some of these groups have hundreds, if not thousands, of bitcoins in reserve, and the depressive effect on the market price is considerable. In effect, it's the likely end of Wall Street speculation on Bitcoin as a commodity, and a return to acting primarily as a currency.

That means Bitcoin will increasingly need to prove itself as a viable currency from now on to survive, but that's also a problem, because of the second reason for the price decline: there are fundamental problems with Bitcoin's infrastructure, and as of yet nobody appears to have a fix for them. It's still far too easy to steal bitcoins, for example - Bitstamp, one of the most popular wallet services, had $5.6m worth of customers' bitcoins stolen just last week, forcing a temporary halt on withdrawals. There's no way to get stolen bitcoins back, either.

Then there's the incoming rent-energy-return on investment crisis in mining. Bitcoin's meant to work like this: the record of every single transaction is the blockchain, which is updated every ten minutes by "nodes", or computers running dedicated Bitcoin software. Every time someone sends someone else bitcoin, that's noted by a nearby node, which then sends further copies out to other nodes, and so on and so on, until as many nodes as possible have a record for when the next blockchain update has to be finalised. If a transaction is recognised by 50 per cent+1 of the nodes on the network, it's legit, and goes through; if it isn't, it's ignored, and that (in theory) means there can't be falsified financial records on the blockchain even though there's no central entity checking every single one.

There are also a predetermined number of bitcoins in circulation, increasing at a set rate - so as an incentive to get people to run nodes 24 hours a day, seven days a week, the new coins coming in get given to the node-runners. Bitcoin transactions are encoded with some complex cryptography, and keeping track of them all requires some decent computing power. It's like trying to crack codes, over and over again, with increasing difficulty, and it's known as "mining".

What's happened over the years, however, is the professionalisation of mining, to an astonishing degree - there are companies selling dedicated mining rigs, and other companies running huge mining operations in basements or warehouses. Vast, hot computers running at top speed, only kept from melting down by equally vast fans. It's a business that comes with a significant cost when it comes to electricity bills and rent. That was fine when the price of a bitcoin was $1000, or $600, or even $400, but as the price of a bitcoin falls it makes those bills inherently more expensive. And Bitcoin is intentionally built so that the difficulty of mining goes up as time goes on, and the number of bitcoins distributed via mining goes down.

In theory, there's a transaction fee on every Bitcoin transaction that's meant to be redistributed to every node owner, and that's in turn meant to compensate for the decrease in the number of mined bitcoins - but again, anecdotal reports seem to indicate that those aren't enough for many miners. And without miners, the bureacratic backbone of the site - the blockchain - becomes more vulnerable. 

Similarly, one of the main ways users try to keep costs low is by ganging together into "mining pools", working together and splitting the profits. Yet this comes with yet further issues: some mining pools have managed to grow so large that they briefly threaten to constitute more than 50 per cent of the nodes on the network, meaning that anyone who managed to control every computer within the pool could, if they wanted to, completely fabricate a new blockchain, and with it give themselves as many coins as they want. If they did it subtly, without being obvious and causing a panic, they could milk the system for a far greater return than any honest mining could provide.

These issues add up to what we're seeing now - the slow, inexorable decline in the price of a digital currency with no value beyond the trust in the system. Where it bottoms-out will depend on how much trust the community has built up among retailers and users up to this point.

Ian Steadman is a staff science and technology writer at the New Statesman. He is on Twitter as @iansteadman.

Show Hide image

“Like a giant metal baby”: whether you like it or not, robots are already part of our world

For centuries, we have built replacements for ourselves. But are we ready to understand the implications?

There were no fireworks to dazzle the crowd lining the streets of Alexandria to celebrate Cleopatra’s triumphant return to the city in 47BC. Rather, there was a four-and-a-half-metre-tall robotic effigy of the queen, which squirted milk from mechanical bosoms on to the heads of onlookers. Cleopatra, so the figure was meant to symbolise, was a mother to her people.

It turns out that robots go back a long way. At the “Robots” exhibition now on at the Science Museum in London, a clockwork monk from 1560 walks across a table while raising a rosary and crucifix, its lips murmuring in devotion. It is just one of more than 100 exhibits, drawn from humankind’s half-millennium-long obsession with creating mechanical tools to serve us.

“We defined a robot as a machine which looks lifelike, or behaves in lifelike ways,” Ben Russell, the lead curator of the exhibition, told me. This definition extends beyond the mechanisms of the body to include those of the mind. This accounts for the inclusion of robots such as “Cog”, a mash-up of screws, motors and scrap metal that is, the accompanying blurb assures visitors, able to learn about the world by poking at colourful toys, “like a giant metal baby”.

The exhibits show that there has long existed in our species a deep desire to rebuild ourselves from scratch. That impulse to understand and replicate the systems of the body can be seen in some of the earliest surviving examples of robotics. In the 16th century, the Catholic Church commissioned some of the first anthropomorphic mechanical machines, suggesting that the human body had clockwork-like properties. Models of Jesus bled and automatons of Satan roared.

Robots have never been mere anatomical models, however. In the modern era, they are typically employed to work on the so-called 4D tasks: those that are dull, dumb, dirty, or dangerous. A few, such as Elektro, a robot built in Ohio in the late 1930s, which could smoke a cigarette and blow up balloons, were showmen. Elektro toured the US in 1950 and had a cameo in an adult movie, playing a mechanical fortune-teller picking lottery numbers and racehorses.

Nevertheless, the idea of work is fundamental to the term “robot”. Karel Čapek’s 1920s science-fiction play RUR, credited with introducing the word to the English language, depicts a cyborg labour force that rebels against its human masters. The Czech word robota means “forced labour”. It is derived from rab, which means “slave”.

This exhibition has proved timely. A few weeks before it opened in February, a European Parliament commission demanded that a set of regulations be drawn up to govern the use and creation of robots. In early January, Reid Hoffman and Pierre Omidyar, the founders of LinkedIn and eBay respectively, contributed $10m each to a fund intended to prevent the development of artificial intelligence applications that could harm society. Human activity is increasingly facilitated, monitored and analysed by AI and robotics.

Developments in AI and cybernetics are converging on the creation of robots that are free from direct human oversight and whose impact on human well-being has been, until now, the stuff of science fiction. Engineers have outpaced philosophers and lawmakers, who are still grappling with the implications as autonomous cars roll on to our roads.

“Is the world truly ready for a vehicle that can drive itself?” asked a recent television advert for a semi-autonomous Mercedes car (the film was pulled soon afterwards). For Mercedes, our answer to the question didn’t matter much. “Ready or not, the future is here,” the ad concluded.

There have been calls to halt or reverse advances in robot and AI development. Stephen Hawking has warned that advanced AI “could spell the end of the human race”. The entrepreneur Elon Musk agreed, stating that AI presents the greatest existential threat to mankind. The German philosopher Thomas Metzinger has argued that the prospect of increasing suffering in the world through this new technology is so morally awful that we should cease to build artificially intelligent robots immediately.

Others counter that it is impossible to talk sensibly about robots and AI. After all, we have never properly settled on the definitions. Is an inkjet printer a robot? Does Apple’s Siri have AI? Today’s tech miracle is tomorrow’s routine tool. It can be difficult to know whether to take up a hermit-like existence in a wifi-less cave, or to hire a Japanese robo-nurse to swaddle our ageing parents.

As well as the fear of what these machines might do to us if their circuits gain sentience, there is the pressing worry of, as Russell puts it, “what we’re going to do with all these people”. Autonomous vehicles, say, could wipe out the driving jobs that have historically been the preserve of workers displaced from elsewhere.

“How do we plan ahead and put in place the necessary political, economic and social infrastructure so that robots’ potentially negative effects on society are mitigated?” Russell asks. “It all needs to be thrashed out before it becomes too pressing.”

Such questions loom but, in looking to the past, this exhibition shows how robots have acted as society’s mirrors, reflecting how our hopes, dreams and fears have changed over the centuries. Beyond that, we can perceive our ever-present desires to ease labour’s burden, to understand what makes us human and, perhaps, to achieve a form of divinity by becoming our own creators. 

This article first appeared in the 23 March 2017 issue of the New Statesman, Trump's permanent revolution