Evidence of technology’s influence in politics is everywhere. In most cases, it is presented as a clear and imminent threat to the foundations of democratic institutions. From fears of election-hacking and data-surveillance to the impact of Facebook ads and weighted algorithms in Google search results, everywhere you look it seems as if emerging data technology is being credited with the downfall of civilization.
A.I. in politics promises to de-facto reinstate a more direct and representative democracy by strengthening the bond between government officials and the people.
Of particular concern is artificial intelligence, which is expanding our capacity to collect and interpret data at an astounding rate. Alarmist reactions to new science are as old as time. Our world must grapple with the legitimate challenges and dilemmas introduced by A.I. and machine learning, but we should also pursue the benefits of these new capabilities. While the infiltration of technological advancement into politics is undeniable, what often goes unrecognized is the immense potential for these applications to strengthen our democratic values and enhance the operations of government.
Artificial Intelligence is transforming the world in dramatic ways and is poised to bring even more sweeping changes that will affect all fields of human endeavor–from social interactions and healthcare to international relations and military operations. Politics is no exception; the advent of A.I. marks the beginning of a completely new era in one of the oldest human fields and, apart from apparent negative repercussions, it will yield tremendous positive uses.
[caption id="attachment_180721" align="aligncenter" width="1920"]
Artificial Intelligence[/caption]
GETTING DIRECT WITH DEMOCRACY
For thousands of years, critics of liberal democracy have argued that it does not represent true “democracy” where citizens make political and economic decisions. Indeed, when we say that countries are “democratic,” we mean that they are “representative republics” because democracy in its original meaning (the direct rule by people) was only practiced in ancient Athens.
Increasingly, politicians study people’s perspectives using A.I.-powered data analytics and then accommodate their views to those of the citizens in order to gain more supporters.
The main problem that arises in representative republics is that representatives often backtrack from their promises and fail to advance the interests of their constituents—or even choose instead to become lobbyists for special interests. Thus, opponents of today’s societal system argue, this inherent gap between government officials and citizens means that representative republics do not further the interests of the democratic majority.
However, the application of A.I. in politics promises to de-facto reinstate a more direct and representative democracy by strengthening the bond between government officials and the people. Increasingly, politicians study people’s perspectives using A.I.-powered data analytics and then accommodate their views to those of the citizens in order to gain more supporters. Barack Obama’s 2012 presidential campaign, Donald Trump in 2016, and Narendra Modi in the 2014 Indian elections, are all prime examples of politicians who successfully employed this strategy.
One of the greatest talents a politician can possess is the ability to attract voters by demonstrating his or her interest in them and genuinely trying to understand their worldview. Such politicians are rare; however, with the use of machine learning in elections, every individual’s opinion will receive attention with A.I. algorithms connecting voters and politicians directly. A.I. will bridge the gap between representatives and citizens and people will have their views taken into consideration, enhancing the democratic process.
In the words of Pedro Domingos, a renowned A.I. expert, thanks to A.I., “democracy works better because the bandwidth of communication between voters and politicians increases enormously.”
[caption id="attachment_180715" align="aligncenter" width="1920"]
Technology[/caption]
DATA-DRIVEN POLICYMAKING
A.I. will help us make better decisions in politics and economics. The use of machine learning, data science, and predictive analytics in governance could enable policymakers to pursue evidence-based policy, with A.I. providing a precise image of what a country needs and how problems could be solved. For instance, by utilizing predictive analytics, A.I. experts can identify potential vulnerabilities in the economy that could trigger a recession and advise on how to tackle them. Indeed, in an AI-powered economy, it might be possible to accurately predict and minimize the negative effects of business cycles.
A.I. could make the process of political decision-making less “political” and partisan and more evidence-based—potentially reducing polarization by forging consensus on issues that simply need to be investigated more thoroughly in order to reach a compromise.
Data-driven policymaking will make the democratic system work better by empowering the government to be more responsive to citizens and provide public services in the most effective ways possible. Instead of relying on personal experience, irrational instincts, misguided dogma, or harmful biases and prejudices, decision-makers could rely on data and A.I. algorithms.
As Daniel Esty and Reece Rushing write, “New information technologies make possible—and affordable—a series of monitoring opportunities, data exchanges, analytical inquiries, policy evaluations, and performance comparisons that would have been impossible even a few years ago… Policymaking, as it currently stands, can be like driving through a dense fog in the middle of the night. Large data gaps make it difficult to see problems clearly and chart a course forward.” By harnessing A.I. technologies, we can close data gaps that have been hindering effective, data-driven decision-making.
A.I. could make the process of political decision-making less “political” and partisan and more evidence-based—potentially reducing polarization by forging consensus on issues that simply need to be investigated more thoroughly in order to reach a compromise.
An intrinsic characteristic of every liberal democracy is the ability to learn from mistakes and adapt to changing conditions. In that, open societies function similarly to the “reinforcement learning” paradigm of machine learning, wherein software continuously learns from its environment to constantly improve itself. Applying advanced reinforcement learning algorithms to politics would considerably increase the efficiency of the government. After all, the more data is available, the better the existing algorithms function, enabling data-driven policymakers to pursue more effective governance.
Open societies have historically been more successful than closed societies because of the former's intrinsic efficiency when it comes to information processing. Centralized states use centralized data processing, whereas open societies are dispersed with multiple data processing units. Distributed processing has proven to be a more efficient model—if one of the multiple processors (government, NGOs, businesses, civil society) failed, others were quick to identify and correct the mistake. In dictatorships, there is a single decision-maker. If this decision-maker is a talented and smart leader (such as in post-Maoist China), the country may enjoy some success but even a minor mistake could entail disastrous consequences because there will be no one to correct it.
Soon, however, the balance of favor might tip towards centralized systems—after all, algorithms work better with more data, and in the centralized system, all data is concentrated in one processor.
It follows then that open societies, as they will be facing competition from authoritarian states, will make decision-making more centralized to gain efficiency–or at least all data will be collected by the government, which will then consult with private companies and citizens in order to make better choices.
[caption id="attachment_180716" align="aligncenter" width="1920"]
Technology[/caption]
FROM SOCIAL STUDIES TO SOCIAL SCIENCES
Perhaps the greatest promise of A.I. is its potential to make social “sciences” true sciences. The main problem with social sciences (such as politics and economics) is that it is hard to make reliable predictions because of the difference between the actual state of affairs and people's interpretation of the world.
If JPMorgan announces that it has created an A.I. algorithm that can predict the behavior of the stock market with 98% accuracy and then regularly tweets its forecasts, the behavior of the stock market itself will change. Conversely, the discovery of Newton’s laws or Einstein’s theory of relativity would not affect the laws of the universe at all.
For example, in astronomy, if we observe the behavior of stars, we are only observers, and therefore our acquisition of knowledge does not affect on the course of events; stars don’t move because we watch them. If, however, the World Bank revises its economic growth predictions, then the reality itself will change as market participants change their behavior, and the prediction might become inaccurate.
Another example: if JPMorgan announces that it has created an A.I. algorithm that can predict the behavior of the stock market with 98% accuracy and then regularly tweets its forecasts, the behavior of the stock market itself will change. Conversely, the discovery of Newton’s laws or Einstein’s theory of relativity would not affect the laws of the universe at all.
Social sciences can’t resolve this fundamental flaw: the inability to factor uncertainty that results from people’s biases, prejudices, and misconceptions. However, the advent of A.I. could enable social scientists to better account for human behavior. After all, large financial organizations already use algorithms to predict stock market moves and asset prices. Considering the breathtaking pace at which A.I. technology progresses, I believe that A.I. will enable us to turn social disciplines into true “sciences”. In the past, due to social sciences’ inherent indeterminacy, it was hard to propose definitive and accurate solutions to existing challenges (for instance, to what extent use monetary and fiscal policies in fighting the recession, or which option to choose when a political adversary makes an unexpected move).
Accomplishing this feat would mark the most important milestone in the history of social sciences. By eliminating uncertainty, we will be able to finally reach consensus on complex political issues and economic problems, promote social cohesion and stability, and help us come closer to achieving humanity’s ultimate ideal of the world—for the greatest obstacle to addressing global problems is the inability to facilitate common attitude and find precise solutions.
A.I., however, will change the rules of the game—and this time, in a positive way.
