How Generative AI Is Changing Job Roles

Harvard just dropped a study on AI and the workforce: "Generative AI as Seniority-Biased Technological Change: Evidence from U.S. Resume and Job Posting Data." It perfectly complements Stanford’s report, published only a a few days earlier. Together, these are the clearest signals yet of how Generative AI is not just changing productivity; it’s reshaping the very architecture of careers. Stanford (ADP payroll data): Since late 2022, employment among 22–25 year-olds in AI-exposed jobs has fallen ~13%, while 35–49 year-olds in the same roles have grown ~9%. Automation-heavy AI uses cut junior jobs; augmentation-heavy ones sustain or even expand them. Harvard (62M workers, 285K firms): At firms that adopt AI (measured via “AI integrator” hires), junior headcount falls 7.7% within six quarters. Hiring slows by ~10% per quarter, even as promotions rise 5%. In Wholesale & Retail, junior hiring contracts by nearly 40%. And graduates from mid-tier universities are the hardest hit. The message is clear: AI is shrinking the base of the career ladder; fewer entry roles, faster promotions for those already inside, and a premium on tacit, senior-level capabilities. The opportunity is differentiation. Companies that design AI-augmented apprenticeships, run talent impact diagnostics, and adopt augmentation-first operating models will not only protect their pipelines but also build the next generation of leaders faster. It seems like AI isn’t just an efficiency story. It’s a career architecture story. Those who act intentionally now will set the tone for an AI-powered workforce that is leaner, smarter, and more resilient. 🔗 Link to Harvard's report ("Generative AI as Seniority-Biased Technological Change: Evidence from U.S. Resume and Job Posting Data"): http://bit.ly/47SyfTC 🔗 Link to Stanford's report ("Canaries in the Coal Mine?"): http://bit.ly/45Ttgzo

The biggest AI impacts won’t be borne out in a calculus of jobs but rather in seismic shifts in the level of expertise required to do them. In our article in Harvard Business Review, Joseph Fuller, Michael Fenlon, and I explore how AI will bend learning curves and change job requirements as a result. It’s a simple concept with profound implications. In some jobs, it doesn’t take long to get up to speed. But in a wide array of jobs, from sales to software engineering, significant gaps exist between what a newbie and an experienced incumbent know. In many jobs with steep learning curves, our analysis indicates that entry-level skills are more exposed to GenAI automation than those of higher-level roles. In these roles, representing 1 in 8 jobs, entry-level opportunity could evaporate. Conversely, about 19% of workers are in fields where GenAI is likely to take on tasks that demand technical knowledge today, thereby opening up more opportunities to those without hard skills.   Our analysis suggests that, in the next few years, the better part of 50 million jobs will be affected one way or the other. The extent of those changes will compel companies to reshape their organizational structures and rethink their talent-management strategies in profound ways. The implications will be far reaching, not only for industries but also for individuals and society. Firms that respond adroitly will be best positioned to harness GenAI’s productivity-boosting potential while mitigating the risk posed by talent shortages.   I hope you will take the time to explore this latest collaboration between the The Burning Glass Institute and the Harvard Business School Project on Managing the Future of Work. I am grateful to BGI colleagues Benjamin Francis, Erik Leiden, Nik Dawson, Harin Contractor, Gad Levanon, and Gwynn Guilford for their work on this project. https://lnkd.in/ekattaQA #ai #artificialintelligence #humanresources #careers #management #futureofwork

Early-career software engineers face an existential job market and it's no longer a debate. Since ChatGPT released two years ago, employment for software developers under 25 has dropped nearly 20%, yet headcount for those over 30 has held steady or climbed. A new paper out of Stanford by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen suggests this is a structural and lasting shift. Generative AI automates much of what used to be done by entry-level programmers. New grads who used to fill junior developer roles are now competing for a fast-shrinking number of jobs. Interesting, too, is just how insulated older, more experience engineers have been through the same period. AI has put a premium on "on-the-job" wisdom. Companies are hiring for the tacit knowledge — judgement, management, people skills — that comes from experience rather than credentials or raw technical ability. It's no coincidence that this divergence started when ChatGPT arrived. It was the first domino to mass adoption of AI, and all evidence points to an acceleration rather than a reversal. To be sure, this opens up new opportunities for the young people who are most fluent in AI. Plenty of 19-year-olds have launched products and startups that have pushed real innovation and led to massive paydays. Still, that doesn't change the takeaway from the data. Junior developer roles are disappearing.