New York Fed Data Shows AI Leads to Retraining, Not Mass Layoffs

Artificial intelligence is driving a massive wave of corporate retraining rather than the widespread job destruction many workers feared. According to a September 2026 regional business survey by the Federal Reserve Bank of New York, 61 percent of service firms and 51 percent of manufacturers are now using AI, yet only 4 percent of service firms and zero manufacturers reported laying off workers because of the technology.[3][4]

The data directly challenges the narrative that generative AI will immediately hollow out the white-collar workforce. Instead of replacing staff, 34 percent of surveyed service firms and 22 percent of industrial manufacturers in the New York and Northern New Jersey region are actively upskilling their existing employees to use the new tools.[2][4]

"Despite this rapid adoption, regional firms' investments in AI are generally modest, usage tends to be concentrated among a small share of workers within firms, and layoffs have remained uncommon," the Federal Reserve Bank of New York noted in its September 1 release. "Retraining employees in response to AI remains the primary way firms are adjusting their workforces."[4][5]

The surge in adoption over the past 12 months has been steep. In 2025, just 40 percent of service firms and 26 percent of manufacturers reported using AI in the New York Fed's district. By August 2026, those figures had jumped by roughly 20 and 25 percentage points, respectively, marking one of the fastest technological integrations in recent corporate history.[2][3]

Yet the employment impact has been remarkably muted. While 15 percent of service firms acknowledged hiring fewer workers than they otherwise would have because of AI efficiencies, another 13 percent actually hired more workers specifically to help implement and manage the technology.[3][4]

Research from Goldman Sachs, released concurrently on September 3, 2026, corroborates the central bank's findings on a global scale. After examining 11 international surveys covering more than 800 occupations, Goldman Sachs Research concluded that aggregate layoffs remain statistically negligible.[2]

However, the transition is not entirely frictionless. The Goldman Sachs report highlights that while incumbent workers are being protected and retrained, entry barriers for junior talent are rising visibly. Generative AI is increasingly handling the basic data-entry and processing tasks that traditionally served as the training ground for recent graduates.[2]

"Junior workers may face stronger headwinds to hiring due to AI adoption," the Goldman Sachs researchers noted, emphasizing that the technology primarily dampens the creation of new entry-level openings in highly exposed sectors rather than destroying existing mid-level roles.[2]

Worker sentiment reflects this shifting landscape, though anxiety remains high. A separate December 2025 survey by the Federal Reserve Bank of Boston found that the share of workers concerned about losing their jobs to AI had doubled from 5 percent in 2024 to just over 10 percent.[6]

The reality on the ground suggests those fears may be outpacing corporate actions. Firms report that they are retraining workers across the educational spectrum, with a slight emphasis on those holding college degrees. The training programs primarily address prompting, software literacy, and everyday task efficiency rather than the creation of entirely new job titles.[2][4]

The consensus emerging from the September 2026 data is that artificial intelligence functions as a tool for task substitution rather than wholesale employee replacement. Employers are utilizing modern applications to augment their incumbent workforce, increasing overall productivity while relying on the institutional knowledge those workers already possess.[1][2]

As the integration deepens heading into 2027, the primary challenge for the labor market will be managing the pipeline for career starters. With mass layoffs failing to materialize, the focus shifts to how companies will onboard and develop the next generation of professionals when their traditional first-year tasks are automated.[2]