Is it really Technology Favors Tyranny?: Notes on Harari's (2018) Article
Artificial intelligence (AI) could erase many practical advantages of democracy, and erode the ideals of liberty and equality. It will further concentrate power among a small elite if we don’t take steps to stop it. - Yuval Noah Harari
I. The Growing Fear of Irrelevance
There is nothing inevitable about democracy. For all the success that democracies have had over the past century or more, they are blips in history. Monarchies, oligarchies, and other forms of authoritarian rule have been far more common modes of human governance.
Harari thinks that the technological revolutions now gathering momentum will be the hardest trials democracy has yet encountered, and that most people in Birmingham, Istanbul, St. Petersburg, and Mumbai are only dimly aware of the rise of AI and its impact on their lives.
But artificial intelligence is different from earlier machines. In the past, machines competed with humans mainly in manual skills. Now they are beginning to compete with us in cognitive skills. And we don’t know of any third kind of skill-beyond the manual and the cognitive-in which humans will always have an edge.
AlphaZero, Google’s chess AI, was one of the strongest AIs and demonstrated superhuman play after only a few hours of training. When people play exceptionally creative moves rather than classical ones, those moves may be suspected as cheating because creativity is increasingly associated with computers rather than humans.
One example is self-driving cars: they often behave like a coordinated system on the road because they run similar algorithms. Humans are separate agents and think differently from each other, but machines can act more uniformly, which can make them safer in some scenarios. Another example is medicine: if the World Health Organization identifies a new disease or a lab produces a new treatment, it cannot instantly update every doctor. But if billions of AI-powered medical systems exist, that knowledge can be distributed with a single update. Even if some humans continue to outperform machines individually, the technologies that make many people economically redundant can also make them easier to monitor and control.
Just as 20th-century governments established massive education systems for young people, 21st-century governments may need large-scale reeducation systems for adults. But will that be enough? Change is stressful, and the hectic early 21st century has produced a global epidemic of stress. As job volatility increases, will people be able to cope? By 2050, a "useless class" might emerge-not only from a shortage of jobs or a lack of relevant education, but from insufficient mental stamina to continue learning new skills.
The Rise of Digital Dictatorships
There is no particular reason to believe that AI will develop consciousness as it becomes more intelligent. We should instead fear AI because it will probably obey its human masters and rarely, if ever, rebel. AI is a tool and a weapon unlike any other humans have developed; it will almost certainly allow the already powerful to consolidate their power.
Harari argues that AI has two major risks for humankind:
- This technology can make many people economically irrelevant.
- It can make people easier to monitor and control.
Consider surveillance. Numerous countries around the world, including several democracies, are building unprecedented systems of surveillance. For example, Israel is a leader in surveillance technology and has created in the occupied West Bank a working prototype for a total surveillance regime. Today, when Palestinians make a phone call, post on social media, or travel between cities, they are likely to be monitored by microphones, cameras, drones, or spyware. Algorithms analyze the gathered data, helping security forces pinpoint and neutralize what they consider potential threats.
These types of control systems are spreading. We have also seen laws and identity systems introduced across Europe and other countries that increase the reach of surveillance, such as ID checks and facial recognition on social platforms.
We are unlikely to face a rebellion of sentient machines in the coming decades, but we might have to deal with hordes of bots that know how to press our emotional buttons better than our family members do. These bots, acting at the behest of a human elite, can be used to sell products, politicians, or entire ideologies.
The current state of Twitter/X illustrates this problem: organized bot networks influence political discourse and manipulate emotions. Some responsibility for this lies with platform decisions and their owners.
As we rely more on Google and large language models for answers, our ability to locate and evaluate information independently diminishes. Today, "truth" is often defined by the top results of a search or by a model's responses. This process has also affected practical skills, like navigation: people increasingly ask a service to guide them rather than learning to find their own way.
We now ask questions of LLMs and often accept their answers without critical evaluation. That reduces independent thinking and the ability to choose among information sources. Harari may have been overly optimistic about the timeline, but his concerns remain relevant.
Harari suggests that AI will change decision-making processes. Societies might grant AI systems permission to make life-changing decisions-about medical treatments, career choices, or even marriage-raising serious ethical and political questions.
More practically, if we want to prevent the concentration of wealth and power in the hands of a small elite, we must regulate data ownership.
I agree with Harari on this point. We should also examine concepts like technofeudalism to understand how economic and social power might be reorganized.
Harari does not fully explain why a digital dictatorship would emerge. Examined topic by topic, his warnings are persuasive, but the causal mechanism for the rise of a digital dictatorship could use further elaboration.
References
Reference: Harari, Y. N. (2018, September 13). Yuval Noah Harari on why technology favors tyranny - The Atlantic.