Governor Kathy Hochul just ordered a massive automated audit of New York's state government rules. The state is partnering with tech nonprofits and academic labs to scan thousands of pages of text. The goal is simple. They want to find every outdated regulation, every redundant fee, and every rule that forces you to use a fax machine.
It's a huge shift in how governments handle bureaucracy. For decades, updating state regulations required rooms full of lawyers working for years. It was slow. It was expensive. Now, machine learning algorithms are doing the heavy lifting in a fraction of the time.
But this isn't just about saving paper. It's an aggressive move to clean up a creaky state system that relies on technology from the 1980s and 1990s. If it works, it could change how every state in America operates. If it fails, it will become another expensive lesson in tech overpromise.
Inside the Regulatory Reset
The initiative is officially called the Regulatory Reset. It kicked off with an executive order designed to target the hidden friction points that slow down daily interactions with state agencies. Think about the last time you dealt with a state agency. You probably had to mail a physical document, provide a wet signature, or get something notarized.
These rules aren't just annoying. They cost small businesses millions of dollars in lost productivity. They clog up state operations.
The state brought in heavy hitters to execute this plan. They are working with the Recoding America Fund and US Digital Response. These are nonprofits that focus on building state capacity and improving service delivery. They also brought in the Stanford RegLab, an interdisciplinary research team packed with engineers, data scientists, and social scientists.
The AI tools aren't making law. They're acting as supercharged researchers. The algorithms crawl through thousands of state regulations, codified statutes, and administrative mandates. They look for specific patterns of inefficiency.
What the Algorithms Are Looking For
The tech teams trained their models to flag specific administrative triggers. The system is scanning for keywords and compliance pathways that don't make sense in 2026.
First, they look for outdated transmission requirements. Any regulation that mandates faxing, physical mailing, or submitting multiple paper copies gets flagged.
Second, the models track down wet signature requirements. In an era of secure digital authentication, requiring a physical ink signature is a massive bottleneck. The software isolates these requirements so agencies can replace them with digital alternatives.
Third, the AI is cataloging every fee and fine imposed on individuals and small businesses. The state wants to find fees that cost more to administer than they actually bring in. If a ten-dollar fee requires thirty dollars of state labor to process, it's a net loss for taxpayers.
Finally, the Stanford RegLab team is using models to audit mandatory reports, boards, commissions, and advisory councils. Over the decades, New York has created hundreds of these entities. Many haven't met in years. Others produce reports that nobody reads. The AI digests these statutes and compiles them into a clean list for agency heads to review.
The Workforce Already Uses These Tools
This isn't New York's first experiment with automation. Earlier this year, Chief Information Officer Dru Rai expanded access to secure generative AI tools for the entire state workforce. That's over 100,000 state employees with access to an internal tool called AI Pro.
A pilot program before the rollout showed interesting results. Three-quarters of the participants saved time on routine tasks. Ninety percent said they understood the technology better after using it. They generated over 170,000 prompts during the trial phase.
Employees use it to summarize massive legal briefs, translate complex policies into plain English, and catch duplicate data across different agency files. When workers use automated tools to handle low-level administrative summaries, they have more time for complex cases that require human judgment.
But using AI to write an email is very different from using AI to audit the law. The Regulatory Reset is a much bigger bet.
The Friction Between Regulation and Innovation
New York's sudden embrace of automated efficiency sits in weird contrast with its legislative track record. The state has some of the most aggressive tech-regulating lawmakers in the country. Just a few months ago, legislators in Albany introduced more than 180 bills targeting automated systems, algorithmic tools, and data centers.
Some lawmakers want a three-year moratorium on new data center construction. They worry that hyperscale facilities consume too much electricity, strain local water supplies, and drive up utility rates for regular consumers.
Then there's the political pushback from labor. The Legislative Oversight of Automated Decision-making in Government Act, or the LOADinG Act, protects public sector jobs. It explicitly states that automated systems can't lead to a transfer of duties usually done by state employees.
This creates a delicate political line for the governor. The state must use tech to eliminate red tape without triggering union contract disputes or massive pushback from environmental advocates worried about the power grid.
Why Vague Claims Won't Fix Government
Tech companies love to promise that automation solves everything. It doesn't. If you feed bad data or poorly structured laws into an algorithm, you get bad recommendations.
The real test of the Regulatory Reset isn't the data scan. It's the human follow-up. The AI creates the map of inefficiencies, but state agency experts still have to review every single flagged rule. They have to decide what to cut and what to keep.
In June, the state announced an initial batch of 50 modernization actions across 22 agencies. That's a drop in the bucket compared to the thousands of opportunities the AI has already flagged. The bottleneck isn't the technology anymore. It's the speed of bureaucratic change.
Actionable Next Steps for Other States
If you're a policy leader, business owner, or local official looking at New York's experiment, you shouldn't just watch from the sidelines. You can apply these principles to your own organization right now.
Map your own workflows before buying tools. Don't look for software until you know exactly where your bottlenecks are. Ask your team what routine tasks consume their time. If your staff spends hours copying data from one system to another, that's where you start.
Isolate transmission bottlenecks. Audit your own processes for mandatory printing, physical signatures, or scanning. Replacing one paper step with a secure digital form can cut processing times by half.
Emphasize human oversight from day one. New York succeeded in its pilot because they treated the tools as an assistant, not a replacement. Every automated summary or rule analysis requires a qualified professional to verify the output. Never let an algorithm make the final policy decision.
Start small with measurable targets. Don't try to rewrite your entire organizational rulebook at once. Pick one department, scan for one specific issue like redundant fees, and fix that first. Measure the hours saved before expanding the program.
