How AI helps the tech giants

Artificial intelligence (AI) and its related technologies — machine learning and the metaverse — represent a watershed in the evolution of the global economy. Like other such shifts, its emergence is likely to favor certain interests, notably a handful of technology giants, the media and a small cadre of highly skilled programmers. Everyone else faces economic danger, certain to roil domestic and international politics in coming years.

Eighty-two percent of millennials fear AI will reduce their earning ability — and they are right to be worried. The first group to lose will be the usual suspects: factory and warehouse workers as well as professionals with largely routinized occupations suited to automation. Service jobs are particularly vulnerable, especially such positions as executive assistants and office managers, long dominated by women.

The most politically disruptive development may derive from the loss of sizable numbers of skilled professionals. Tech firms such as Salesforce, Meta, Amazon and Lyft have announced major cutbacks in their white-collar workforce and have warned that these positions are unlikely to return. IBM has put hiring on hold while assessing how many mid-level jobs can be replaced by AI. Google has recently laid off 12,000 workers, a number that is expected to grow to 30,000. The damage may be even greater at the grassroots level. Within months of AI’s emergence, freelance work in software declined markedly, along with pay for the jobs that remained.

Yet artificial intelligence presents a great opportunity for the economy as a whole. PricewaterhouseCoopers estimates artificial intelligence technologies will add $15.7 trillion to the global economy by 2030. But this boom will likely be more feudal and stratified than earlier tech waves. The “early digital idealists,” notes technology analyst Jaron Lanier, envisioned a “sharing” web that functioned “free from the constraints of the commercial order.”

In contrast, the AI revolution is fostering dependent small satrapies that serve the existing giants of the industry. This new configuration helps those who can tap enormous financial interests such as pension and sovereign wealth funds, who have provided upwards of $7 trillion in capital for new high-end chips and the development of ever more complex and sophisticated algorithms, even as global cash for startups is at the lowest ebb in five years.

AI likely will accelerate the shift towards corporate giantism. Already, Google and Apple account for nearly 84 percent of all mobile browsers worldwide and Microsoft and Apple operating systems control 89 percent across all desktops and laptops. A relative handful of large digital platforms also dominate the $421 billion digital advertising market. Meta, Google, Amazon, X (formerly Twitter), TikTok and Alibaba being the major players globally. Perhaps more ominously, two-thirds of the world’s cloud services — essential for AI and the operation of most digital servers — are controlled by Amazon, Microsoft and Google.

The very logic behind AI, its reliance on existing records and databases, is not ideal for startups; its “primary value,” notes venture capitalist Martin Casado, is “to improve existing operations for incumbents who have the resources to invest at the required levels.” AI may spark improvements in education, medicine and even infrastructure design and maintenance, but the odds that smaller companies will play large developmental roles are slim. Big-tech executives such as LinkedIn and Inflection co-founder Reid Hoffman promise that AI will serve the cause of “elevating humanity,” reflecting the “techno-optimism” embraced by venture capitalist Marc Andreessen. Yet the impact on employment may not be so utopian. Some projections have AI wiping out hundreds of millions of jobs worldwide. In the US, according to McKinsey, at least 12 million will be forced to find new work by 2030.

It’s obvious that AI and enhanced machine learning will accelerate the loss of blue-collar jobs. Warehouse workers will be among the most prominent losers. This extends also to people taking digital orders; Walmart expects to automate its systems with new software and lay off 2,000 workers by 2026. The push for AI-driven automation will be critical in the future, particularly in countries such as Japan and Germany, with their rapidly aging workforces.

AI could also threaten the social and medical services which have experienced huge growth in recent decades. Tech firms are looking to develop “something like your personal AI”; others are developing new robotic nannies. There are already bots that duplicate the work of professionals: by harvesting his total oeuvre into cutting-edge AI software, students of prominent psychologist Martin Seligman came up with a prototype chatbot that Seligman agrees gives much the same advice he would. Less intellectually demanding services may get the AI treatment too, if the need for human sex workers is outsourced to bots. Will the world’s oldest profession disappear?

AI might be most disruptive to the very professional classes that once benefited most from digitization. A recent survey suggests that two-thirds of business leaders agree that ChatGPT will soon lead to large layoffs of white-collar workers, including coders and symbolic analysts. “We may be at the peak of the need for knowledge workers,” Atif Rafiq, a former chief digital officer at McDonald’s and Volvo, told the Wall Street Journal last year. “We just need fewer people to do the same thing.”

Already new AI programs allow software firms to do without lower-level programmers. Even practiced geeks who write code may be vulnerable to what economists refer to as “skills-biased technological change.”

Not even the movies are safe from potential AI “plagiarism machines” — you don’t need Shakespeare to write the next Marvel movie, or its sequels.

So, who survives the onslaught? Some elite AI engineers could experience a windfall; big global investors could become much richer. Surprisingly, skilled manual workers like mechanics and oil-riggers may be difficult to replace with robots. According to AI pioneer Rony Abovitz, the big winner in the coming years will be the “sophisticated, technically capable blue-collar worker.” Goodbye to “my son the computer geek,” hello to “my son, the plumber.”

All this economic upheaval is certain to disrupt our culture and politics. Already roughly three in five Americans, according to a recent poll, see AI as a direct threat to civilization. Former Google executive Geoffrey Hinton, an early developer of AI, has warned that its capacity to create convincing false images and texts will create a world where people will “not be able to know what is true anymore,” a phenomenon that seems primed to exacerbate the already mounting unease that has encouraged the populist right in both Europe and North America.

But problems may not come only from the right: if AI begins to affect elite livelihoods, they too could become radicalized. Though people displaced in Cambridge or Palo Alto are not likely to become followers of Donald Trump, oligarchs’ support for progressive green or gender policies may not make up for the blow to elite bank accounts.

Support for replacing work with guaranteed income may therefore grow stronger: Tech writer Gregory Ferenstein recently determined that many digital-company founders believe an “increasingly greater share of economic wealth will be generated by a smaller slice of very talented or original people. Everyone else will come to subsist on some combination of part-time entrepreneurial ‘gig work’ and government aid.” And roughly half of all Americans support a guaranteed basic income of about $2,000 a month if robots put them out of work.

But ultimately such socialistic trends may mean people — left or right — balk at allowing tech oligarchs to hold on to their current power and massive assets. Some progressives envision a future where tech and Wall Street wealth is confiscated to fund “fully automated luxury communism” — a kind of tech-powered leisure society.

The rise of AI is bound to affect global politics. The titanic competition between the United States and China, and their occasionally symbiotic technological push-pull, is unlikely to vanish overnight, and prominent AI advocates such as Sam Altman and key companies like Nvidia generally seek to accommodate their Chinese partners. This inspires considerable political worry, and much demand for controls on AI along with other technology. But if political hurdles discourage Western competitiveness, the likely effect will be to allow Asian countries, notably China and other autocratic regimes like Saudi Arabia, to “rule the world by 2100,” in the words of a recent Brookings Institution study.

Some westerners feel that China’s authoritarian structures will stunt the country’s technological rise but given its progress, this seems a bit overblown. China’s AI firms, with strong backing from the government, are already finding ways around sanctions. The Middle Kingdom is not standing still; it’s capitalizing on its leads in surveillance and new media algorithms.

Defense applications of AI have grown explosively in the past decade, as we see with the use of AI drones in the Ukrainian conflict. They are also being used in the Red Sea and in and around Israel, and are (somewhat ironically) a threat to Taiwan, where most advanced AI chips are now being developed. The fact that North Korea is developing its own AI capacity should keep military planners awake at night.

AI works off the surveillance of people and ideas; whether or not it’s benignly applied, it’s intrusive by nature. And China, with assistance and funding from western tech firms, is already showing how advanced digital technologies can be used to control society through total surveillance, most oppressively in Xinjiang. The fact that China now exports such systems to sixty-three countries, particularly in the developing world, is concerning.

AI’s enhanced ability to craft images, shape narratives and monitor preferences are certain to be felt in Western societies, no matter what political controls are proposed. Privacy is already an illusion; virtually all of our activities are tracked and recorded, then dropped into a vast digital stew of databases. The implications are stunning: can AI help bosses keep close watch on their employees, even outside their jobs? What happens if electronic health records, or financial ones, are mined? These dangers suggest governments, supported by research institutions, might look for ways to limit AI’s impact. Yet many citizens are wary of ceding control; handing power to the state has inherent problems, particularly given the lack of diverse views inside the tech community and the universities.

There are likely to be even greater challenges to resolving the upheavals caused by artificial intelligence. An increasingly alienated populace, hostile both to ever more intrusive government and to the increasing concentration of power and riches in ever fewer hands, may demand a reckoning. In promoting AI, the tech oligarchs, their financial backers and political allies may gain vast new wealth, but also reap some unpleasant consequences associated with an unprecedented disruption in the lives of average folk.

This article was originally published in The Spectator’s April 2024 World edition.