What Bitcoin And Novel Coronavirus Have In Common

On getting up to speed on 2019-nCoV, I observed early the resemblance of its mathematics to a 2019 paper using Metcalfe’s law to value bitcoin. As both are exponential growth models encompassing the network effect, this was unsurprising. However, when I thought longer, the similarities appeared more profound:

• Both show an increasing trust deficit between the Chinese public and its leadership

• Both reveal china’s vulnerability is not economic but political

• A comparison shows the limits and dangers of China’s longer-term AI strategy

Information and Trust

Why Bitcoin is so popular in China is keenly debated – some say it is for technical reasons (cheap electricity and ascendant hardware), and some political. The strength of bitcoin over periods of Chinese market stress suggests that political reasons play at least some part. Even without ideological intent, rational actors under a system of tight exchange control express distrust by using bitcoin as a practical means of wealth conservation. Bitcoin is attractive in this environment because it challenges the information asymmetry inherent under authoritarianism.

Without doubt, nCoV is President Xi’s greatest test since coming to power and remaking the Chinese Communist Party (CCP) in his image. Questions about the government’s control of information are central to current criticism, and even those of us outside of China can see signs of public discontent. The prime example is the outpouring of grief over Dr Li Wenliang’s death from nCoV – he was one of the first to report the new strain and police subsequently apprehended him for disturbing public order. Paucity of live information is not unusual in these viral outbreaks; many authorities faced criticism over the H1N1 outbreak in 2009, including the US Centers for Disease Control and Prevention (CDC). The key difference, however, is that most don’t arrest their doctors. Information is power and, when concentrated, bad outcomes lead to a breakdown in trust.

Owning Outcomes

An anonymous member of the Hive Community likes to say that trust is in a bear market. It is hard to disagree as authoritarians all over the world are finding ways to exploit this. Their methods can be broadly split into two. Democracies have long given up on one true narrative. Instead they use social media to flood the public with a confusing mass of information. This information is often emotionally charged and designed to come from a competing array of platforms all dressed in the appearance of equal credibility. This is the censorship through noise strategy. By contrast, traditional authoritarianism insists on one clear truth and tightly filters any dissenting voices. This is the one narrative strategy.

The key difference in these models is the ownership of outcomes. Referencing Nassim Taleb, we could say that censorship through noise has anti-fragile properties (an increased capability to thrive as a result of stressors) while the one narrative does not. Censorship through noise relies not on legitimacy but kinship and loyalty. Resultantly, these leaders can acknowledge bad outcomes with less ownership and fewer adverse consequences for themselves. By contrast, one narrative leaders cannot evade ownership and are more vulnerable to bad outcomes.

Economically, I am a China bull and have never believed the country to be as economically vulnerable as some in the West promote. Chinese authorities still wield extraordinary power to manage economic outcomes. While unfashionable, it still appears to me that China’s vulnerability is political. Controlling data when you can manage the outcome is very different to controlling data when you cannot. NCov is China’s first non-economic shock for some time, and consequently China’s one narrative leadership is under pressure we have not seen since the last political paradigm reversal. Deng Xiaoping’s reform efforts in the 1980s focused on the decentralization of power with the decade’s unrest related to those who wanted to take it too far, too quickly. In current times, China is in the process of culling decentralized nodes of power and returning that power to the centre. While modern technology with its face recognition and overwhelming surveillance capabilities appears to suit these efforts, a more nuanced consideration of AI and big data point in the opposite direction.

AI and Big Data

The idea that AI could lead to human extinction is borderline consensus in the academic and tech world. The reasons why and the scenarios in which this could happen are many, but a central theme is that writing optimization code for opaque implementation channels will eventually have unintended consequences. The classic example is telling an AI to eradicate spam email and it eventually concludes that the best way to do so is to eradicate emails, which is best done by eradicating humans. While the future may not prove anywhere near as dreary, the principle is clear: AI at scale risks unintended consequences. At some point, something big will go wrong with AI systems. The bigger, more complex, and more socially focused these systems become, the greater the chance of political ramifications. While this could occur anywhere, China’s one narrative government will find it particularly difficult to explain away.

A more prosaic but potentially larger danger comes from big data. Big data requires big networks, and the challenge of securing these grows along with the size of the network. Indeed, the chance of data breaches, leaks, or just plain errors becomes a real headache for those charged with protecting the network. If these challenges are severe enough in cyberspace, the probability of mishap only increases as these nodes interact with the physical world in hospitals (physician’s terminals), on street corners (video cameras), or in the hands of citizens (smartphones). With the impossibility of knowing what sort of content may prove explosive, one narrative systems are always vulnerable to the discovery and dissemination of information counter to the official stance.

The Bottom Line

To be clear, political change in China is not imminent and the pressures faced by the incumbent government are still far from existential. Moreover, once the severity of the nCoV situation was recognized, authorities have been quick to act, enacting measures – such as quarantining a population larger than California – that would be unimaginable elsewhere. However, this ongoing and tragic episode does reveal an alternative, non-economic scenario for how political change may occur in the future. As China’s focus on government-sponsored AI leadership increases, the chances of a technology driven black swan event that leads to political and social upheaval increases along with it.