The Economist: AI won’t kill white-collar jobs — it will transform them

Warnings about artificial intelligence wiping out white-collar jobs have been piling up. Kristalina Georgieva, head of the IMF, likened AI’s impact on the labour market to a “tsunami,” while Jamie Dimon, CEO of JPMorgan Chase, predicted his bank would soon require fewer employees. Dario Amodei of Anthropic went further, forecasting that AI could erase “half of all entry-level white-collar jobs.” Yet, a closer look at the data and historical trends suggests that while AI is transforming office work, it is far from rendering it obsolete, according to The Economist.

Since late 2022, the U.S. has added roughly three million white-collar jobs across management, professional, sales, and office roles, while blue-collar employment has largely stagnated. In sectors often cited as vulnerable—software development, radiology, and paralegal work—employment has risen by 7%, 10%, and 21%, respectively. Real wages have held steady, with professional and business services salaries up 5% and administrative workers seeing a 9% increase. White-collar workers now earn roughly a third more than blue-collar ones, a premium nearly triple that of the early 1980s. Far from undermining office work, AI has so far coincided with continued prosperity in the sector.

This pattern mirrors earlier technological revolutions. The early computer era prompted dire predictions about mass displacement, yet it ultimately expanded white-collar employment. Computers automated routine, repetitive tasks but rarely replaced entire roles. Professionals used automation to focus on higher-value activities—analysis, judgment, and decision-making—rather than being displaced. The same holds for AI today. Current systems excel at discrete tasks—drafting text, coding, or data analysis—but struggle with the “edge cases” and discretion that define most professional jobs. Research from Anthropic suggests that only a small fraction of occupations can be automated across three-quarters or more of their tasks.

Evidence from the past three years reinforces this perspective. Roles requiring technical expertise combined with oversight—project managers, cybersecurity experts—and those demanding interpersonal care or judgment have grown rapidly. By contrast, routine back-office positions, such as insurance claims clerks and administrative assistants, have declined by 13–20%. AI is also creating new job categories, from data annotators to “chief AI officers,” with emerging roles often commanding rising wages.

Nonetheless, disruption is inevitable. Entry-level roles and tasks with few discretionary elements are most at risk. AI models capable of sustained autonomous work, including coding and analysis, may soon encroach on jobs previously considered secure. Yet historical experience and current labour-market trends suggest that AI is more likely to reshape roles than eradicate them. Human judgment, accountability, and adaptability remain key drivers of value.

In short, AI is transforming the office into a cyborg environment—part human, part machine. White-collar workers may need to adapt and upgrade their skills, but the end of office work is far from imminent. Instead, we are entering an era where human creativity, oversight, and discretion complement AI’s capabilities, potentially raising productivity, wages, and opportunities rather than eliminating them. The Economist’s analysis underscores a reassuring message: AI may redraw jobs, but it won’t erase them.

By Vugar Khalilov