AI Is Coming for All of Our Jobs

Yet, even while most top executives had up until this point been cautious when speaking publicly, the Journal revealed that behind closed doors, they had already ​“spent months whispering about how their businesses could likely be run with a fraction of the current staff” with the aid of artificial intelligence and robotics.

For the founders of the San Francisco-based start‑up company Mechanize, even the decimation of half the white-collar workforce is not sufficiently ambitious. Mechanize, which has been funded by investments from top Silicon Valley insiders, including Patrick Collison, CEO of the payments company Stripe, and Jeff Dean, Google’s chief scientist, aims to displace all human labor, with the goal of ushering in ​“a fully automated economy.” The objective, says 29-year-old co-founder Tamay Besiroglu, is to ​“make that happen as fast as possible.”

Mechanize’s website points out that the potential riches from accomplishing this are ​“absurdly large” given that ​“workers in the U.S. are paid around $18 trillion per year in aggregate,” while ​“for the entire world, the number is over three times greater, around $60 trillion per year.” This rapidly emerging concern (or enthusiasm) about the prospect of wiping out the laptop class is, of course, the result of the astonishing advances in AI systems employing large language models (LLMs) that first captured the world’s attention when OpenAI’s ChatGPT was introduced in November 2022. The remarkable progress in these generative AI systems has been driven by intense — often brutal — competition between the world’s most powerful technology companies.

The latest capabilities of LLM-based systems from OpenAI (backed by Microsoft), Google parent Alphabet, Anthropic (backed by Amazon and Google), Meta, and Elon Musk’s xAI are relentlessly compared and ranked, with the results displayed on leaderboards and continuously scrutinized by AI researchers, engineers and investors. This real-time gamification of the race between the top AI offerings has served to further accelerate the already astonishing levels of investment in both the compute infrastructure and the human talent required to compete in the AI race.

In June 2025, for example, Mark Zuckerberg — reportedly panicked by the perception that Meta was falling behind OpenAI and Google — offered up to $100 million to some of OpenAI’s top employees and ultimately succeeded in luring away at least three key researchers. Zuckerberg’s desperation was perhaps justified in light of the fact that many top AI experts believe the industry might achieve true human-level AI — known as Artificial General Intelligence (AGI) — within roughly the next five years. Indeed, many in the field are already looking beyond AGI and anticipating the near-term arrival of artificial superintelligence. The total investment powering this competition is staggering. Alphabet, Amazon, Meta, and Microsoft were estimated to have spent $410 billion in 2025 alone.

So far, the actual revenue generated from the AI systems the companies have built doesn’t come close to justifying that level of investment. OpenAI, for example, estimated that it would lose about $5 billion in 2024, as the cost of the compute infrastructure to operate ChatGPT, together with the near-astronomical salaries for its engineers and researchers, greatly exceeds its revenue. Needless to say, losses like this are not sustainable indefinitely. Eventually, the industry will have to generate a return on all that investment. And the prospect of saving corporate customers untold billions by deploying AI to automate jobs almost certainly looms large as the tech companies search for a profitable business model.

To be sure, the founders of Mechanize are not the only ones who have noticed that the potential market for AI systems capable of automating much of the work now performed by salaried workers is ​“absurdly large.” Accelerating progress in information technology — and especially artificial intelligence and robotics — will dramatically reshape the job market, the economy and society in the coming years and decades. 

Ultimately, these technologies are destined to displace human labor and disrupt our current system to the point that a new social contract and economic paradigm will become inevitable. The transition into this new future will likely pose one of the greatest challenges our society has faced in generations. 

While many of the most accomplished AI experts believe that continuous improvements to today’s major LLM platforms will lead directly to AGI in less than a decade, others are far more skeptical. If the critics are right, progress may plateau, and breakthroughs in entirely new research areas may be required before Artificial General Intelligence can be achieved. However, the automation of a wide swath of both white-collar and blue-collar work does not require machine intelligence that equals or exceeds the human brain in every domain. 

This is because virtually all jobs are specialized. If AI can automate your job — or a large fraction of the tasks that make up your job — then it does not matter if it lacks capability in other, unrelated areas. Indeed, one of the most important recent developments is the emergence of specialized AI tools built on top of the major LLM models. Microsoft, for example, claims that its new AI Diagnostic Orchestrator, a system that can leverage any of the major AI platforms, is up to four times more accurate than human doctors when tasked with making a particularly challenging diagnosis. Similarly specialized systems are emerging in fields like law, finance, customer service and, especially, computer programming.

As both start‑up companies and large corporations develop such systems, we may well see a Cambrian explosion of LLM-based AI applications targeted at specific industries or professions. At the same time, numerous competing engineering teams are racing to develop AI agents — systems that can go beyond answering questions and directly perform the kinds of tasks currently undertaken by human knowledge workers.

The pace of progress in this new domain of agentic AI will likely be one of the most important determinants of the timing of white-collar job displacement. While the potential for chatbots and AI agents to vaporize white-collar work is currently the hottest topic of conversation, the field of robotics is also seeing remarkable progress. Indeed, the latest advances in AI are being used to animate ever more powerful machines capable of physical manipulation.

Amazon has already deployed more than a million robots in its distribution centers, and the machines are soon expected to outnumber human workers. In the company’s most advanced facilities, the robots are already performing tasks — such as grasping varied items from shelves — that just a short while ago would have been achievable only with a human worker’s dexterity and visual perception. As robots continue to improve and become more flexible and likely more affordable, they will be deployed in ever-increasing numbers in warehouses, factories and other similarly controlled environments. But they will also begin to invade fast food restaurants, supermarkets, and other service-sector businesses.

In other words, a wide range of both blue-collar and white-collar work will be increasingly vulnerable to automation. Artificial intelligence is rapidly evolving into a general-purpose technology, and as it powers algorithms and animates physical machines, it is set to disrupt virtually every industry, every profession and every employment sector.

I believe strongly that artificial intelligence will be the single most important technological force that shapes the coming years and decades. The impact on society at large — and on each of us individually — will ultimately be profound.

Excerpted from RISE OF THE ROBOTS: Technology and the Threat of a Jobless Future by Martin Ford, ©2026 by Martin Ford. Used with permission of Basic Books, a division of Hachette Book Group, Inc.