The January 31, 2006 issue of The Economist included a special focus on the mounting anxiety about the social consequences of AI, with four articles devoted to the subject. “Solving fiendish maths problems, making complex medical diagnoses, conjuring up new software in moments: the feats of generative AI get more impressive by the day,” noted the issue’s lead article, urging readers to “Stop panicking about AI. Start preparing.”
While the future course of AI is obviously uncertain, there are good reasons to believe that society has time to prepare and adapt. “It takes time for a new technology to diffuse from the cutting edge to the office cubicle,” the article noted. Firms and governments should use this breathing space to help those most at risk of displacement.
“So far labour markets seem unruffled. Service jobs are most exposed to generative AI, yet in America the number of white-collar jobs has gone up by 3m since ChatGPT was launched, while blue-collar jobs have stayed flat. Employment has risen even in areas that have been keen adopters, such as coding.” One reason for the slow economic impact is that while AI excels at some tasks, it also “confidently spouts nonsense, or struggles to count the number of ‘r’s in ‘strawberry.’” This unpredictability means that companies and workers need time to figure out where, and how, to apply AI effectively.
“Moreover, business processes don’t change overnight. Electricity was first harnessed commercially in the 1880s, but it took 40–50 years to generate productivity gains on factory floors. Plants had to be redesigned and workflows rethought.” This time, too, companies must think carefully about how to encourage workers to use AI, how to mitigate its shortcomings, and how to deploy it successfully.
Realizing the potential of a general-purpose technology (GPT) — like steam power, digital computers, and now AI — requires large investments and a fundamental rethinking of how production is organized. It takes considerable time for these technologies and new business models to be widely deployed across economies and for their full benefits to be realized.
Let me briefly summarize the three other articles in The Economist’s issue.
How Big a Threat Is AI to Entry-Level Jobs?
“The classic shape of the company hierarchy is a pyramid: lots of people at the bottom, tapering to a single point at the top,” notes one article. But, it’s quite possible that firms will need many fewer people because pyramids may become passé in the AI age. If entrepreneurs can create huge firms on their own, the future organization might resemble a dot. But the shape that captures the biggest near-term worry is the diamond: if entry-level and junior jobs are disproportionately affected by AI, white-collar organizations may become narrower at the top and bottom and wider in the middle.
The article references a recent paper on AI’s employment impact, “Canaries in the Coal Mine?”, co-authored by Stanford professor Erik Brynjolfsson and colleagues. The authors found evidence that some firms may indeed be moving toward a diamond-shaped organization, as employment among entry-level young workers in AI-exposed occupations has begun to decline.
In a related Substack post, co-author Bharat Chandar summarized the paper’s key findings:
- Employment among 22–25-year-olds is declining in AI-exposed jobs such as software development and customer service.
- Jobs automated by AI are seeing declining entry-level employment, while jobs augmented by AI are not.
- Overall, the job market for experienced workers remains strong, but entry-level employment has been stagnant.
Other papers caution that AI may not be the only source of economic uncertainty; job prospects for young workers were already weakening before the launch of ChatGPT in 2022.
“There are, however, good first-principles reasons to be worried about AI’s impact on entry-level jobs,” adds The Economist. “Many junior staff spend their early years doing document-heavy grunt work in exchange for experience. Now there is a machine that can do document-heavy grunt work more efficiently and cheaply than humans.” If junior employees are using AI to complete this work, are they actually learning? And if the diamond is the shape of the future, why not let other firms train early-career workers and hire them later?
The article offers three strong reasons why managers should be cautious about slashing entry-level hiring:
- First, no one knows how AI will affect work over the longer term.
- Second, while firms always risk losing trained workers, failing to build a pipeline of future talent can be even riskier.
- Third, cutting junior roles is not the best way to build an AI-literate workforce — especially since younger workers tend to use AI more readily than older ones.
Why AI Won’t Wipe Out White-Collar Jobs
Since the arrival of ChatGPT in November 2022, AI “has excited and terrified in equal measure,” noted The Economist in a second article. The head of the International Monetary Fund warned that AI is “hitting the labour market like a tsunami.” JPMorgan Chase’s CEO predicted that the bank would soon need fewer employees. The leader of Anthropic suggested that AI “could wipe out half of all entry-level white-collar jobs.”
AI could indeed wreak havoc across the white-collar workforce. “But rather than making many such jobs less lucrative, or redundant, it is likelier to reshape them,” argued The Economist. “The AI office will look less like a robot and more like a cyborg, combining the best of human and computer capabilities: the Six Million Dollar Man rather than Terminator.”
“For all the alarm, white-collar workers are still doing well,” the article added. Since late 2022, the U.S. has added roughly 3m white-collar jobs, while blue-collar employment has remained flat. To understand what may come next, it helps to look at earlier technological revolutions.
In a 2015 paper, “Why Are There Still So Many Jobs? The History and Future of Workplace Automation,” MIT economist David Autor explained the impact of technological revolutions on jobs.
“Broadly speaking, many, perhaps most, workplace technologies are designed to save labor,” wrote Autor. Given the steady stream of labor-saving innovations, why hasn’t employment collapsed?
The answer, he argued, lies in the structure of jobs themselves. Most jobs consist of multiple tasks, some of which are easier to automate than others. Automating routine tasks does not eliminate jobs; instead, it often increases workers’ productivity and allows them to focus on judgment, coordination, and interpersonal skills.
“Roles that combine technical expertise with oversight and coordination have enjoyed the biggest gains,” The Economist noted. Employment among project managers and information-security experts has risen by roughly 30%. Jobs requiring deep quantitative expertise, problem-solving, interpersonal care, and judgment are also thriving. “Only routine back-office work has shrunk. Over the past three years or so the ranks of American insurance-claims clerks have shrunk by 13% and those of secretaries and admin assistants by 20%.”
How to Avoid Common AI Pitfalls in the Workplace
The final article opens with an AI experiment at a restaurant in the Dallas area, that Pizza Hut uses as a real-world laboratory. “It is a place where the worlds of melted cheese and artificial intelligence collide.”
Customers place orders by speaking to a voice-enabled AI system, while machine-learning algorithms prioritize kitchen workflows. Given that fast-food restaurants have high staff turnover, AI chatbots also help new employees learn the ingredient quantities they should use for different orders.
AI is spreading “into all corners of the workplace. But it still feels incremental, not transformative.” While AI boosters talk of superintelligence and the end of work, “here on planet Earth, the technology merely increases the odds of getting the right number of pepperoni slices on your next takeaway.” These modest experiments raise important questions: Are the benefits merely incremental? What is holding back progress? And what should firms do to get more value from AI?
Critics may argue that a pizza restaurant is not the best testing ground for advanced AI. But that is another way of saying that AI’s impact is unevenly distributed. While many software-development jobs may be deeply affected, most nursing skills remain beyond AI’s reach.
Although AI models are improving rapidly, adoption will take time. As with past technological revolutions, organizations and workers must adapt. “A prosaic set of management problems needs to be solved,” The Economist concluded, including incentives for adoption, guardrails to manage risks, and systems for selecting and measuring applications. “You need a mixture of pragmatism and ambition.”
The lesson running through these various articles is not that that disruption can be avoided, but that we’re not out of time. Labor markets have not collapsed. Organizational change is slow. General-purpose technologies take decades, not years, to reshape economies. That window should not be wasted. Preparing workers, rethinking job design, investing in complementary skills, and experimenting thoughtfully with AI will matter far more than grand predictions about the end of work.
There is time to adapt to AI. The real challenge is whether we use that time wisely.