New York AI Regulation: How New Laws Could Affect Employers

AI is becoming part of everyday work faster than many employers expected. A company might use AI to sort resumes, screen applicants, write job descriptions, evaluate performance, answer employee questions, or analyze workplace data. That convenience is useful, but it also creates a new legal question: what happens when an employer lets an algorithm influence an employment decision?

In New York, employers need to pay attention because AI regulation is developing at both the state and city level. For businesses in New York City, some rules already apply to automated employment decision tools, while additional proposals and policy changes are continuing to shape the conversation around workplace AI.

New York Does Not Have Just One AI Employment Law

It can be tempting to search for a single “New York AI law” and assume that is all an employer needs to know. The reality is more complicated.

Different rules can apply depending on:

• Where the employer operates
• What the AI system actually does
• Whether it affects hiring or promotion
• Whether employees or job applicants are evaluated
• What type of information the system processes
• Whether the employer is a private company or government organization

New York City is particularly important because it already has specific requirements for certain automated employment decision tools.

NYC’s Automated Employment Decision Tool Rules

If a business operates in New York City, Local Law 144 of 2021 is one of the most important rules to understand.

The law covers certain automated employment decision tools, commonly called AEDTs. These are computer-based systems that use technologies such as machine learning, statistical modeling, data analytics, or artificial intelligence to substantially assist with employment decisions.

For covered tools, employers and employment agencies generally must make sure a bias audit has been conducted before using the tool, make a summary of the audit publicly available, and provide required notices to affected candidates or employees.

That means an employer cannot necessarily buy an AI hiring product, turn it on, and forget about it.

What Is a Bias Audit?

A bias audit is intended to examine whether an automated employment decision tool produces significantly different results for certain groups.

The basic idea is pretty understandable.

Imagine an employer receives 1,000 applications. An AI system reviews those applications and recommends 200 people for further consideration. If the system consistently screens out particular groups at significantly different rates, that could raise discrimination concerns.

An audit does not magically guarantee that an AI system is fair. But it provides a way to examine the tool before relying on it.

NYC’s rules require covered employers to use an AEDT only when the required audit and other obligations have been satisfied.

Employers Also Have Notice Responsibilities

Transparency is another major part of NYC’s AEDT requirements.

Employers using covered tools must notify job candidates and employees that the tool will be used to assess them. The notice also includes information concerning how someone can request a reasonable accommodation.

For employers, this means the AI system shouldn’t be treated like a completely invisible part of the hiring process.

A candidate should not have to wonder months later whether an algorithm rejected their application without anyone ever explaining that automated technology was involved.

AI Can Create Traditional Employment Law Problems

Even when a new AI-specific law does not directly apply, traditional employment laws can still matter.

For example, suppose a company uses AI to rank applicants. The employer may believe the system is objective because a computer is making the recommendation. But if the system relies on flawed historical data, it could potentially reproduce discriminatory patterns.

The fact that a machine made the recommendation does not automatically make the employment decision legally safe.

This is why employers should be careful about treating AI output as unquestionable.

Human Review Still Matters

One of the simplest protections an employer can put in place is meaningful human review.

That doesn’t necessarily mean an employee has to manually redo every task the AI performs. Instead, the level of review should make sense for the importance of the decision.

Using AI to help draft a job description is one thing.

Using AI to reject an applicant, terminate an employee, determine a promotion, or make a compensation-related recommendation is much more serious.

Employers should know when AI is simply assisting a person and when it is actually influencing the final decision.

Employee Data Is Another Concern

AI tools can process a huge amount of information, which creates another issue for employers: employee privacy.

An employee might enter information into an AI tool without thinking twice. Perhaps they upload a performance review, paste an internal email, or ask an AI system to summarize a document containing personal information.

Before allowing this, employers should understand how the AI provider handles submitted data.

A company should consider creating clear rules about:

• Confidential employee information
• Social Security numbers and financial information
• Medical information
• Passwords and login credentials
• Internal business documents
• Customer information
• Trade secrets
• AI vendors and third-party platforms

The easiest AI tool to use isn’t necessarily the safest one for workplace information.

New York Is Also Watching AI’s Effect on Workers

The discussion goes beyond hiring.

New York policymakers are increasingly looking at how AI affects the broader workforce. A 2026 New York Assembly economic report discussed both the growing demand for AI-related skills and reported AI-associated job losses, while cautioning that the available data does not establish that AI alone caused those employment declines.

New York City has also enacted a 2026 local law requiring a study and report concerning the impacts of algorithmic tools and automated employment decision tools on city employees.

This shows where the conversation is heading: regulators aren’t only asking whether AI can make hiring faster. They’re also asking what happens to workers when algorithms become part of everyday employment decisions.

What Should New York Employers Do Now?

Employers don’t necessarily need to stop using AI.

Instead, they should understand exactly where it is being used.

Start by making a list of the AI systems employees currently use. You may be surprised by what turns up.

Then ask:

1. Does the tool influence hiring or employment decisions?
2. Does it evaluate applicants or existing employees?
3. What employee or applicant information does it process?
4. Has the vendor conducted the necessary testing or audit?
5. Are required notices being provided?
6. Is there meaningful human review?
7. Can employees challenge or correct an inaccurate result?
8. Does the company’s AI policy clearly explain acceptable and prohibited uses?

For employers in NYC, checking compliance with Local Law 144 should be a priority when using a covered AEDT. NYC’s Department of Consumer and Worker Protection continues to provide guidance and enforcement information for businesses.

Frequently Asked Questions

1. Does New York ban employers from using AI for hiring?

No. New York City allows covered automated employment decision tools, but employers using them must follow applicable requirements, including bias-audit and notice rules.

2. Does Local Law 144 apply throughout New York State?

No. Local Law 144 is a New York City law. Employers elsewhere in New York should separately determine which state and local requirements apply to them.

3. Do employers have to tell applicants when AI is being used?

For covered AEDTs in NYC, employers and employment agencies have notification obligations to candidates and employees.

4. Can an employer rely completely on an AI hiring recommendation?

That can create significant risk. Employers should understand what the system is doing and maintain appropriate human oversight, particularly when AI influences consequential employment decisions.

Final Thoughts

AI can make hiring and workplace administration quicker, but employers shouldn’t confuse automation with accountability.

For New York businesses, the important question isn’t simply whether AI is being used. It’s how the technology is being used and whether it affects people’s employment opportunities or rights.

NYC employers using automated employment decision tools already have specific requirements to consider, and the broader discussion around AI and employment continues to develop in 2026.

If your company uses AI for recruiting, hiring, employee evaluation, scheduling, or other important workplace decisions, reviewing the technology and its compliance requirements now can help avoid much bigger problems later. For a specific employment or compliance issue, a New York employment attorney can evaluate the particular AI system and circumstances.