Harvard finds that AI-native startups employ fewer junior-level positions.

      A working paper from Harvard Business School and INSEAD reveals that AI-native startups are 25% smaller, hire 13% more engineers, and have about 15% fewer entry-level employees and managers compared to non-AI counterparts. Their workforce tends to be more senior, highly educated, predominantly male, and based in Silicon Valley, indicating that AI is creating concentration rather than equalizing opportunities.

      According to the research, led by Rembrand Koning and Hyunjin Kim, which analyzed Y Combinator startups from 2020 to 2024 along with other US venture-backed companies, AI-native startups are characterized by two key shifts: using AI to enhance employee productivity internally and integrating AI into products for customers to automate tasks previously managed by human teams.

      The data shows a stark contrast, with AI-native startups being 25% smaller, employing 13% more engineers, and featuring approximately 15% fewer entry-level staff and managers. The proportion of senior employees is 20% higher, and although their valuations are similar to those of non-AI firms, the implication is that they generate more value per employee.

      The composition of the workforce in these firms reveals a tendency toward specific demographics. The researchers noted that “these workers are particularly likely to be graduates from prestigious institutions, located in Silicon Valley, and male.” This observation challenges the optimistic narrative surrounding the AI boom, where it was believed that junior employees would leverage AI to improve their skills and lower barriers to entry in technical fields. Instead, the study suggests that opportunities may be increasingly concentrated among those who are already well-qualified.

      The authors express concern regarding growing inequality, indicating that if AI enhances learning for its users, “differential adoption rates may lead to widening performance disparities.” This applies to employees within companies as well as the entrepreneurs who establish them.

      The findings align with existing trends in the labor market, where AI is diminishing the availability of summer internships and prompting an increase in graduate unemployment. Recent graduates represent only 7% of new hires at leading tech firms.

      Major tech companies are reallocating resources from payroll to computational capabilities, with Meta and Microsoft reducing their workforces by 23,000 positions as AI investments soar. Meanwhile, demand for AI talent is so high that AWS is investing $1 billion in forward-deployed AI engineers.

      The implications of this study are concerning for those portraying AI as a tool for democratization. While the technology may streamline company hierarchies, it could also make entry into these organizations significantly tougher.