Companies investing in AI are hiring

Companies that spend the most on artificial intelligence have increased their workforce, not reduced it. This finding comes from a report by Ramp Economics Lab, published together with Revelio Labs, which cross-referenced data on corporate spending in AI with workforce information from 21,599 U.S. companies.

The document identifies two categories of firms: the "high-intensity adopters," which is the third of companies that spend the most on AI per employee (about $30 per month per person in the first three months, a figure set to rise), and all others. Only the first group shows a measurable effect on employment: in the 24 months following the adoption, their workforce grew by 10.2%. Companies with low spending intensity, those that limit themselves to subscriptions or pilot experiments without ongoing investments, do not record any significant personnel gains.

Surprising data on junior roles

The most surprising data concerns junior roles, which are considered to be most exposed to algorithmic replacement. Among companies that invest the most, entry-level workforce grew by 12%, with an increase of 1.15 percentage points in the share of newly hired juniors compared to the control group. Growth was seen in engineering, sales, administration, customer service, finance, marketing, and scientific roles, with the IT sector (software, internet, media, and adjacent tech companies) leading the way.

Limitations of the research

Ara Kharazian, chief economist at Ramp and author of the study, acknowledges the limitations of the research. The high-intensity adopter companies were already bigger, more engineering-oriented, more often supported by venture capital, and faster-growing than average. This makes it difficult to determine how much of the hiring truly depends on AI and how much simply relates to the fact that these are already expanding companies.

The paper itself admits this directly: "This work does not demonstrate that AI creates jobs universally but disproves the idea that AI will lead to widespread job losses." This distinction contrasts with other recent research: a study by Goldman Sachs estimates that AI has already eliminated about 16,000 net jobs per month over the past year, with effects concentrated particularly on Gen Z workers and entry-level profiles.

A picture of polarization

The picture that emerges from Ramp suggests rather a polarization: those with capital, technical personnel, founder networks, and managerial capabilities to turn AI adoption into structural investment are achieving concrete returns in terms of growth. Those who settle for scattered tests and subscriptions are falling behind, without employment benefits nor, likely, productivity benefits.

The report comes at a time when another widespread narrative in the sector is crumbling, that which claims that open-source models could not generate revenue. Companies releasing open models are starting to record tangible revenues, a sign that businesses are shifting from simple rental of AI services towards executing proprietary models in-house. This is a piece that complicates, rather than clarifies, the debate on how artificial intelligence is really reshaping the labor market.