AI Is Reshaping New York's Small Businesses — But the Risks Are as Real as the Rewards

New York small businesses spent roughly $2.3 billion on AI-related tools and services in 2025, according to figures from the New York City Economic Development Corporation — a fivefold jump from three years earlier. The money is moving fast. The thinking about consequences is not keeping up.

The pressure to adopt is everywhere right now. With inflation still grinding at operating margins, labor costs in the city running among the highest in the country, and larger competitors deploying automation at scale, owners who once swore they'd never touch the technology are signing up for monthly subscriptions to platforms like Salesforce Einstein, Intuit Assist, and a half-dozen others. The question hanging over every Chamber of Commerce meeting from the Flatiron District to Flushing isn't whether to use AI — it's whether anyone fully understands what they're agreeing to.

The Promise Arrives on Flatbush Avenue

The optimism is genuine and, in places, well-earned. At the Brooklyn Navy Yard, where the city's BNYDC incubator houses more than 500 businesses, several tenant companies in food production and logistics have cut scheduling errors by more than 30 percent after adopting AI-driven inventory tools. The city's Small Business Services agency launched a free AI literacy program in February 2026 specifically targeting the 200,000-plus businesses that employ fewer than 20 people — the engine room of the outer-borough economy.

In Jackson Heights, Queens, a cluster of independent retailers along 74th Street began using shared AI customer-service chatbots through a neighborhood business improvement district pilot in March. Early data from the program shows after-hours inquiry response rates improving dramatically. Owners say it lets them compete with chains that have full customer support staffs.

The allure is understandable. Manhattan commercial rent averaged $84 per square foot in the first quarter of 2026. Every hour not spent on administrative work is, effectively, money recovered.

Where the Trouble Starts

But the ethical and practical risks are piling up in ways that city regulators are only beginning to confront. New York's Local Law 144, which requires bias audits for automated employment-decision tools, covers hiring algorithms — but it leaves enormous gaps. AI systems making pricing decisions, flagging customers as fraud risks, or generating marketing copy based on demographic data operate largely without oversight. Advocacy groups at the Urban Justice Center on Broad Street have documented cases in the Bronx and East Harlem where algorithmic pricing tools used by small landlords and service businesses produced outcomes that disproportionately charged lower-income ZIP codes more for equivalent products.

Data security is a separate, compounding problem. Many small business AI subscriptions feed customer data — purchase histories, addresses, behavioral patterns — to third-party servers. A March 2026 report from the Manhattan-based nonprofit Center for Democracy and Technology found that fewer than 12 percent of small business owners who adopted AI tools in the past year had read their vendor's data-sharing terms in full. Most didn't know where their customer data was being stored, or who else had access to it.

Workforce displacement remains the sharpest edge. The Fiscal Policy Institute, headquartered on Park Avenue South, projects that AI automation could affect roughly 340,000 clerical and administrative jobs in New York City by 2028 — concentrated heavily among workers without four-year degrees. A chatbot that saves a Jackson Heights retailer four hours a week is, in aggregate, pressure on a labor market that absorbed those hours as actual employment.

City Council legislation introduced in May 2026 by a coalition of Brooklyn and Queens members would require any business operating in New York with more than 10 employees to disclose AI-assisted customer interactions. The bill is in committee and faces industry opposition.

For business owners trying to navigate all of this right now, the practical starting point is specificity. The city's Small Business Services office on Rector Street offers free one-on-one consultations — as of this month, appointments are available within two weeks. Owners who go in armed with their vendor contracts and a list of what data their tools collect will leave with clearer answers than those who show up asking whether AI is good or bad. That question, at least, already has an answer: it's both, and the details are everything.