Some of the world’s most powerful technology executives are investing heavily in artificial intelligence while simultaneously preparing for the possibility that it could threaten humanity.
From underground bunkers to isolated islands, some of the industry’s wealthiest figures appear to be making plans to shield themselves from the very risks associated with the technologies that their industry is helping to develop.
The benefits of technological innovation are not necessarily shared by everyone, nor are its risks. Those who profit most from developing new technologies may not be the same people who bear the consequences when things go wrong.
Sociologists have long studied this unequal relationship through the concept of “risk positions.” Just as people occupy different positions in society according to wealth and economic power, they also occupy different positions when it comes to risk.
At one end of the spectrum are those who reap substantial benefits from producing risks while avoiding most of their consequences. At the other are those who benefit little from these activities but bear a disproportionate share of the costs.
Who benefits and who pays?
The economic rewards associated with producing risks can be enormous. Industries that generate environmental damage or financial instability, for example, can create substantial private wealth while imposing significant costs on society.
Yet few industries have produced fortunes as extraordinary as the technology sector. Some of the fastest-growing companies in history have emerged from the tech sector, while their founders have accumulated astronomical levels of wealth.
The prospect of accumulating even greater wealth creates a powerful incentive to keep developing risky technologies. But this is only half the story. To understand why risks may be produced beyond what is socially desirable, we also need to consider who pays for them.
When those who produce risks do not bear their full costs, they have an economic incentive to keep producing them, even when the consequences for society are potentially devastating. The more effectively producers of risk can avoid these consequences, the greater the burden placed on everyone else.
In this framework, it’s not enough for producers of risk to benefit from said risks; they also need to be able to avoid any potentially catastrophic consequences.
They need to occupy the positions of “value skimmers” — those who benefit from upside of risk-taking while avoiding its potentially devastating downside. They are distinct from “value surfers” who gain from risks, but also bear the consequences when those risks materialize.
Preparing to escape the consequences
The distinction becomes particularly revealing when considering the growing tendency among tech executives to prepare for potential future catastrophes.
Several prominent figures in the AI industry have publicly warned that advanced AI could pose an existential threat to humanity. Yet those warnings don’t seem to have slowed the development of AI at all. Locked in a battle for AI supremacy, there doesn’t appear to be an inch to give between competing tech giants.
There are also concerns about generative AI that have come from within the industry itself. Employees and researchers working in the field have raised alarms about the scale and ambition of AI development, including comparisons to the Manhattan Project and technologies reminiscent of the fictional Terminators.
Yet while some of the industry’s leaders appear to be doing their best to insulate themselves from these risks, their companies’ business models depend on expanding digital infrastructure and collecting as much data as possible.
This expansion forces ordinary people to be constantly dependent on digital connection and leaves them fewer opportunities to avoid the risks associated.
As AI becomes increasingly integrated into digital networks, those networks also become more vulnerable and risky. To prevent any aspect of our life from being outside internet dependence is to force us to be constantly at the mercy of technology companies.
Could the public push back?
German sociologist Ulrich Beck argued that modern society is one in which we must constantly face the social, political and economic problems of distributing a “poisoned cake.”
His metaphor captured a central problem of what he called the “risk society” — the benefits of modern development are distributed unequally, while the dangers it creates can extend far beyond those who produce them.
Yet Beck did not believe citizens would take this lying down. He argued they would come to see the reality of the risks being produced and act collectively to arrest control of this system away from those producing them.
The conditions for such a moment may now be emerging. Protests against the expansion of data centres alongside parents’ demands that governments protect children from the harms of social media are just two trends that suggest people are coming to recognize their real risk positions.
Perhaps people’s acceptance of bearing the brunt of the risks that they did not create is coming to an end. Whether it’s the people, the politicians or the tech giants who will win this debate is still to be decided.
