/Tech3h ago

Joseph Barrow and Denis Peskoff of UC Berkeley release a dataset of 2.2 million US city and county laws

Story Overview

Two UC Berkeley researchers have pulled together 2.2 million city and county ordinances that were scattered across vendor sites and PDFs into one searchable corpus, turning rules that were technically public but practically invisible into something machines and people can actually query at scale.

36011.8K1.6K5.4K810.6K

#33

Original post

Michael Guimarin@Guimarin

This guy collected every local law in America and put them in a single database.

2.2 million laws.

This might seem to you like some nerds side project, and while it is technically, it is also much more important than you think.

Historically, whenever anything is made transparent to the voting bloc it undergoes radical change.

Every. Single. Time.

To the extent governments work overtime to make things opaque. (And I expect that to happen to this dataset too very soon, in a thousand different ways.)

The internet makes everything including politics “global/local” and local laws won’t escape.

It’s hard to understand right now how profound this will be, but we live in interesting times for sure.

Thank you Joe, and all the other nerds who do stuff like this because they are personally curious and they can. This is how history moves forwards

Joe Barrow@barrowjoseph

New paper: every law in America is technically public. But not really, until now!

With @DenisPeskoff at UC Berkeley, we built a corpus of ~every publicly accessibly city and county law, and released a huge chunk of it!

2.2 million laws, you're (probably) covered in it!

🧵

8:34 PM · Jun 19, 2026 · 290.6K Views

Open Question

Access stays limited to researchers for now

The main release sits on Hugging Face as an open dataset aimed at legal AI work, with a harmonized county layer covering most of the U.S. population, but no public search tool or commercial API has been announced yet.

Developer Impact

Downstream uses remain to be proven

Companion classifiers for topic, function, opacity, and paternalism were released alongside the data, yet real-world studies on how the corpus shifts housing policy, licensing reform, or local enforcement are still ahead.

Sentiment

Many users praised the LOCUS corpus release of 2.2 million US local laws as an impressive tool for transparency and exploration, while others called the volume of laws tyrannical or highlighted access barriers and local corruption.

Pos

70.5%

Neg

29.5%

56 comments with sentiment.

Cluster Engagement

Digg Deeper

No Digg Deeper questions have been answered for this story yet.

HUGGINGFACEVia

#33

Posts from X

Most Activity

VIEWS53.6K

Joe Barrow@barrowjoseph

Once we had processed the laws, the question was: how to analyze them.

One idea I really like is @TheStalwart's http://Havelock.ai, where he distills pairwise comparisons from an LLM + TrueSkill into a BERT model to score orality.

1d53.6K25251

BOOKMARKS369

Joe Barrow@barrowjoseph

Paper Link: https://arxiv.org/abs/2606.19334 Dataset: https://huggingface.co/datasets/LocalLaws/LOCUS-v1 Models: https://huggingface.co/LocalLaws

And obviously a huge thank you to @modal for enabling this work.

1d10.8K417369

LIKES523REPLIES10

Joe Barrow@barrowjoseph

First, the released dataset has pretty good geographic coverage, and accounts for a majority of the US population!

Everything including zoning, noise, housing codes, and whether or not you can ride an ATV without a license.

1d17K52352

RETWEETS730

Joe Barrow@barrowjoseph

New paper: every law in America is technically public. But not really, until now!

With @DenisPeskoff at UC Berkeley, we built a corpus of ~every publicly accessibly city and county law, and released a huge chunk of it!

2.2 million laws, you're (probably) covered in it!

🧵

1d534.8K4.1K2.6K

Joe Barrow@barrowjoseph

How do you build such a dataset when all the laws come in heterogeneous formats? To me the obvious answer is: OCR it all!

In this case, using the smallest/best OCR model we could, @LightOnIO's LightOnOCR-2-1b.

That gives us a high-quality, consistent format we can parse.

1d13.7K39791

Joe Barrow@barrowjoseph

California and Florida, you need to get your shit together so people can actually understand your laws! And Ohio and West Virginia, wtf is going on with how you run people's lives??

1d9.5K41414

Joe Barrow@barrowjoseph

In our case, we do the same for the laws across 4 axes: paternalism, opacity, enforcement discretion (how much leeway does the law provide), and salience.

We analyzed the 2.2 million laws and some interesting patterns emerged.

1d12.8K28426

Joe Barrow@barrowjoseph

Thankfully @modal (and @charles_irl) very kindly provided a compute grant to make this work happen!

You wanna *really* see GPU's go brrr? (And CPUs, for rendering)

We were able to process all 7M pages at ~$0.30/1k, much cheaper than options like AWS Textract ($1.50/1k pages).

1d12.3K28422

Joe Barrow@barrowjoseph

Our released dataset scores all 2.2 million laws along these axes, and we've released the distilled models for scoring/understanding the laws!

All distilled into @benjamin_warner/@bclavie/@antoine_chaffin/@orionweller/@answerdotai's ModernBERT.

1d10.2K23919

Joe Barrow@barrowjoseph

N.B. I wonder if we're the first academic citation for @TheStalwart's @Havelock_AI.

If there's a better way to cite, let me know and I'll update the preprint!

1d52.1K21018

Joe Barrow@barrowjoseph

First approach was to try and make my local GPUs go brrrr.

That worked quite well for the first million pages, but we had millions more to go, and needed to scale.

1d11.7K2517

Joe Barrow@barrowjoseph

@trallala93 @DenisPeskoff Really hoping to release something soon for this!

1d2.8K885

matt duffy@iammattduff

@barrowjoseph @DenisPeskoff this is incredible! i've been trying to build a similar dataset for zoning code for months. i don't think your ocr-it-all method will work for my use case, but just putting this together is awesome. i'm excited to dig into the data

18h4.8K1042

Joe Barrow@barrowjoseph

@iammattduff @DenisPeskoff If you give me more info about the shape of the data happy to offer advice!

17h4.3K1001

Joe Barrow@barrowjoseph

@Guimarin > some nerds side project

😭

14h2.9K772

Joe Barrow@barrowjoseph

@matospiso @DenisPeskoff Thanks! Was 7M pages total, I and I think the released set comes from around 2.5M pages.

1d3.4K562

JMS@jms_usmc

We are drowning in laws citizens couldn’t read if it was our full time job while the state continues to insist “ignorance of the law is no excuse”. We need legislation garbage collection. My idea is simple. All legislation expires in a few years and must be read aloud in normal speaking cadence on the floor of the voting body to be voted on again. This limits the number of laws they can keep on the books.

6h591371

champagne papineau@trallala93

@barrowjoseph @DenisPeskoff What do you think it would take to get these laws into a queryable/analyzable form? Like suppose I’m a researcher and I want to understand if a certain ordinance affects something I measure. How does one approach this?

1d3.3K431

AK@_akhaliq

Freeing the Law with LOCUS

A Local Ordinance Corpus for the United States

54m2K83

Jscott@J_Scott_Art

@barrowjoseph @DenisPeskoff Do you think thus could be compared against how often these laws are actually applied and enforced? Like could you also cross reference every court case, fine, permit rejection, and arrest? Thatd give a view of not just what the law intends but what it actually does in practice

7h407133