A Harvard Business School and INSEAD working paper finds AI-native startups are 25% smaller, employ 13% more engineers, and carry roughly 15% lower shares of entry-level workers and managers than non-AI peers. Their hires skew senior, elite-educated, Silicon Valley-based, and male, suggesting AI is concentrating rather than democratising opportunity.TL;DR
Startups built around AI hire fewer entry-level workers than their peers, according to a working paper from Harvard Business School and INSEAD, first reported by Business Insider. The firms are leaner, flatter, and heavily weighted towards senior technical talent.
Researchers Rembrand Koning and Hyunjin Kim examined Y Combinator startups from 2020 to 2024 alongside a broader set of US venture-backed firms. They define AI-native startups by two shifts: using AI internally to make employees more productive, and embedding it in products so customers can automate work that once required human teams.
The numbers are stark, with AI-native startups 25% smaller, employing 13% more engineers, and carrying roughly 15% lower shares of both entry-level workers and managers. The share of senior workers runs 20% higher, and valuations are comparable to non-AI peers, implying more value created per employee.
The workers these firms do hire skew a particular way. “These workers are especially likely to be graduates from elite institutions, concentrated in Silicon Valley, and male,” the authors wrote.
That cuts against the hopeful reading of the AI boom, in which juniors use AI to punch above their grade and vibe coding lowers the technical bar. The paper suggests opportunity is instead concentrating among the already credentialed.
The authors’ deeper worry is compounding inequality, warning that if AI accelerates learning for those who use it, “differential adoption rates may translate into widening performance gaps”. That applies to workers within firms and to the entrepreneurs who found them.
The bottom rung is cracking
The findings echo what is already visible in the labour market, where AI is killing the summer internship and graduate unemployment is climbing. Recent graduates now make up just 7% of new hires at major tech companies.
Big Tech is busy converting payroll into compute, with Meta and Microsoft cutting 23,000 roles as AI spending hits records. Demand at the top is so hot, meanwhile, that AWS is putting $1bn into forward-deployed AI engineers.
Even hiring itself has become an AI-on-AI arms race. For new graduates, the machines now sit on both sides of the table.
The study’s implication is uncomfortable for anyone selling AI as a democratising force. The technology may flatten hierarchies inside companies while steepening the climb to get into them.