NY Fed survey reveals most firms invest modestly in AI tools

For all the noise around artificial intelligence and its potential to upend the labor market, a fresh read of the latest New York Federal Reserve data reveals a quieter, perhaps more telling reality: most businesses are dipping their toes into the AI pool rather than diving in headfirst.

The survey of regional firms in New York and Northern New Jersey shows AI adoption is soaring—but the actual financial commitment remains surprisingly shallow, suggesting a cautious, experimental phase rather than a full-blown industrial revolution.

The numbers tell a story of mass adoption on the surface. Over 60 percent of service firms now report using AI, a sharp jump from 40 percent a year ago.

Among manufacturers, usage has doubled to 51 percent from 26 percent in 2025. Yet when it comes to putting money behind this technology, most firms are holding back.

Three-quarters of service firms and over 90 percent of manufacturers describe their AI investments as "minimal to modest," ranging from free tools to a small slice of overall spending.

Only 15 percent of service firms have committed significant resources, and a mere 5 percent view AI as a major strategic play. No manufacturers at all have made such a leap.

This cautious investment pattern is key to understanding the broader labor impact.

While the public debate often fixates on job destruction, the Fed's data paints a picture of a workforce being adjusted rather than gutted.

Only 4 percent of service firms reported layoffs due to AI over the past six months, up from just 1 percent last year, while manufacturers reported zero layoffs for the second year running.

Our analysis suggests this low layoff figure is not an accident but a direct reflection of the limited scale of AI deployment.

When firms invest only modestly and confine AI usage to a median of 17 percent of service workers and 7 percent of manufacturing workers, the disruption to overall headcount is naturally contained.

The real workforce adjustment is happening through hiring decisions rather than pink slips.

About 15 percent of service firms said they hired fewer workers than they would have otherwise due to AI, but this was partially offset by 13 percent of service firms that actually added headcount to help leverage the technology.

For manufacturers, the hiring reduction effect was minimal with none reporting increased hiring. This creates what one could call a "hiring substitution" dynamic.

The survey also underscores the primacy of retraining.

More than a third of service firms and over a fifth of manufacturing firms using AI are retraining their existing workers, focusing on everything from basic AI literacy to prompt engineering and responsible use protocols.

What does this mean for a small business owner or an employee watching these trends? Our reading of the data suggests several actionable takeaways.

For businesses considering AI, the barrier to entry is lower than often assumed.

The Fed survey notes that cost is not a primary deterrent for non-adopters; instead, about half say their work doesn't lend itself to AI, and a quarter feel AI is not good enough yet.

This points to a practical strategy of starting small with free or low-cost AI tools for routine tasks—marketing copy, data analysis, or accounts payable—rather than waiting for a perfect solution.

The "human-in-the-loop" oversight mentioned by some firms in the survey is a smart starting point.

For employees, the message is clear: the safest path is to become the person who can use AI effectively. The data shows retraining is the dominant response, not replacement.

Investing personal time in learning to verify AI outputs, understand bias, and apply tools to your specific job function could make you more valuable, not less, as your firm experiments with adoption.

Looking ahead, three years of survey data suggest the "augment not replace" narrative holds for now. The Fed analysts conclude that AI has been more likely to reshape work than eliminate jobs.

However, this cautious reality could shift. The same report notes a recent study suggesting entry-level workers may be more vulnerable, as AI substitutes for routine tasks often assigned to new hires.

The patterns seen today are not destiny.

As one Fed analysis puts it, "AI technology and its applications are still evolving rapidly, and these patterns could shift as adoption matures." The modest investments of today may be the foundation for bolder moves tomorrow, but for now, the data offers a tempered view of how businesses are actually using AI—one measured step at a time.

Source: Jaison R.

Abel, Richard Deitz, Natalia Emanuel, and Nick Montalbano, "Businesses Are Using AI to Transform Work, Not Cut Jobs," Federal Reserve Bank of New York Liberty Street Economics, September 1, 2026.