A new analysis from the Federal Reserve Bank of New York finds that small businesses currently using artificial intelligence tools are significantly more likely than non-users to expect increased employment and revenues over the next twelve months. The finding matters because it offers a rare forward-looking view of how AI adoption may shape labor demand at the firm level, even as economy-wide measures of pay and headcount have moved little. The results suggest that, at least among small firms, AI use is associated with expansionary expectations rather than workforce contraction.
The analysis draws on the 2025 Small Business Credit Survey, which found that 46 percent of firms with at least one employee reported that their business or employees were using AI tools, with another 15 percent planning to adopt them within twelve months. Among current AI users, 63 percent said AI was somewhat or very important to production, and 51 percent had integrated AI partially or fully into business processes. Despite this adoption, 77 percent of respondents reported no change in current labor costs from AI use, indicating that near-term cost effects have been muted even as expectations shift.
The employment expectation gap is substantial. Among firms with at least one employee, current AI users had a 33 percentage point net expectation of higher employment over the next twelve months, compared with only 15 percentage points for non-users. The gap narrows but remains statistically significant after controlling for owner and firm characteristics, including age, profitability, firm size, and location, at 13 percentage points. It persists at 10 percentage points even after accounting for past employment changes. Notably, the association is stronger among firms reporting fair or poor current financial conditions, where AI users show a 21 percentage point higher net probability of expected employment growth, versus 10 percentage points among firms in good or better condition.
The revenue picture is more tempered. Only 31 percent of survey respondents reported increased sales from their firm's use of AI. However, revenue expectations are most optimistic for AI users that report facing operational challenges in utilizing technology, suggesting that firms grappling with integration difficulties may still anticipate meaningful gains. The analysis also finds that among AI users, those who have fully or partially integrated AI and those who regard AI as important to production are equally likely to expect higher employment relative to those merely experimenting with AI, after accounting for firm and owner characteristics.
The evidence is limited to a single survey wave and relies on owner-reported expectations rather than realized outcomes. The authors note that firm expectations could be explained by unobserved characteristics, though they attempt to address this using financial condition as a proxy for optimism and find the opposite of a spurious correlation. What remains unknown is whether these expectations translate into actual hiring and revenue growth, and whether the pattern holds for larger firms or across different AI use cases. Future survey waves and matched administrative data would help determine whether the optimism documented here is durable or fades as adoption matures.