This article raises difficult but important questions about the role of technology in human society, and points to AI/ML as a potential threat to social cohesion, quality of life, and human dignity and freedom. This is a topic of special importance to me, and Harari's piece will serve as the starting point for a multi-part discussion of technology, politics, and humanity.
The issues raised in Harari's piece are disturbing to say the least. He begins with the admonition that, "there is nothing inevitable about democracy." The current state of politics in the US and worldwide is a stark reminder of this inevitability - we see the norms and institutions of democracy increasingly threatened. It is difficult to deny that 'liberal democracy' as we have known it in the last several decades is facing a crisis of credibility and relevance.
Against this backdrop of increasing political and economic upheaval, technology is ever more rapidly disrupting conventional institutions and practices. In the big picture of human history, there is a fundamental disconnect between the promise of technology and its actual impact on human society. Keynes famously predicted a 15-hour work week for his grandchildren, but that vision obviously has not come to pass. With this in mind, Bill Gates' suggestion that we should not fear AI because it will result in longer vacations seems downright preposterous.
Technology has undeniably increased human productivity, but the benefits of that productivity have accrued almost entirely to corporations and to a small number of wealthy individuals, and not to workers in general. The AI revolution promises to exacerbate this trend of economic and political inequality. As Harari says, "the same technologies that might make billions of people economically irrelevant might also make them easier to monitor and control."
There are two competing visions for the coming revolution - one of technology as a savior, a catalyst for enabling mankind to ascend to a higher level of material, moral, and intellectual development. The other is of technology as a nightmare, a tool of alienation, oppression, and exploitation.
I would argue that, for all the growing awareness of the perils of technology in recent decades and years, there is still a strong cognitive bias towards its advantages and benefits, and away from its costs. New technology often offers solutions to the problems created by previous rounds of technology, and new advancements happen so quickly that technologies often achieve widespread adoption before it is possible to have a real social conversation about the possible costs, and to look at the big picture of how to integrate new technology with existing ideas, structures, norms, etc.
More on this soon - here is a review of Harari's book Homo Deus: A Brief History Of Tomorrow that I wrote for a machine learning class.