It's not false economy. The time it takes to generate a 50-state survey by hand is far greater than the time it takes to verify Lexis/Westlaw's AI answers and correct the errors.

An update for Sonnet 4.5, released last week: it scored 60.2% on my final exam (with extended thinking on, 54.4% without it). That's a big step up (~20 percentage points) from Opus 4.1's scores, and puts Sonnet 4.5 close to, if slightly behind, other lead models. On a human curve, that's ~ an A-/B+

One other note: across the five exam answers and dozens of answer evaluations generated here, I did not notice a single hallucination. This test wasn't designed to measure hallucination rates, but it's consistent with the general sense that they have dropped significantly

Our office is again hiring one or more attorneys for a one-year fellowship to work directly with the Illinois Solicitor General and her team, beginning in August/September 2026. www.governmentjobs.com/careers/ilag...

For my latest round of informal tests of large language models, I looked at how good different models are at taking a law school exam—and also whether they are capable of grading exam answers in a consistent and reasonably accurate way. 🧵 www.wilftownsend.net/p/chatgpt-ta...

It was very nice to have two of my recent articles featured in JOTWELL reviews this month—Maureen Carroll on "Deterring Unenforceable Terms," courtslaw.jotwell.com/should-draft... and @margotkaminski.bsky.social on "The Deletion Remedy" cyber.jotwell.com/ai-disgorgem...

A nice quick read from my colleague @JonahPerlin about an issue that I see a lot of people oversimplifying: whether an attorney's use of a generative AI tool waives privilege. This is an area where I'm very interested to see how the law develops. news.bloomberglaw.com/us-law-week/...

In a stunning moment of self-delusion, the Wall Street Journal headline writers admitted that they don't know how LLM chatbots work.

A very pleasant surprise to listen to one of my favorite podcasts and hear my own work being discussed. And it's an excellent episode and overview for anyone thinking of AI's effects on the legal profession. Some thoughts / suggestions below for anyone who wants further reading:

Judge Alsup has the first true opinion on fair use for generative AI in Bartz v. Anthropic. He holds that AI training is fair use, and so is buying books to scan them, but that downloading pirated copies of books for an internal training-data database is not fair use. 🧵

I think this is one of the more common mistakes I see with people trying AI—the idea that if you go to a free chatbot, quickly run a question by it, and it does a bad job, then you've learned that AI cannot do a good job on that question.

Significant ruling in one of the big algorithmic price-fixing lawsuits going on right now: www.reuters.com/legal/govern...

A good thread on a big new generative AI / IP lawsuit—Disney and Universal vs. Midjourney

ChatGPT is down but The Museum of English Rural Life still stands, proving once again that Silicon Valley cannot compete with the history of rural England and its people.

I've had a few recent conversations with judges and law profs who haven't tried generative AI, or have only used it for a few minutes to write a poem or other trivial fun. After a few people asked me about how to start, I thought I'd write up my suggestions: www.wilftownsend.net/p/some-ideas...

After years of studying AI & law here’s my rule of thumb: Lawyers should only use AI only when they can confidently assess, adapt & explain its output without engaging in deep, independent thinking about the core legal/factual issues. More in my new essay: papers.ssrn.com/sol3/papers....

This is one of my favorite books to recommend to students (I actually have two copies). One of my teaching aspirations is to someday have a seminar read this alongside @nbagley.bsky.social’s Procedure Fetish and Kagan’s Adversarial Legalism.

"Litigation...is not simply a tool...it embodies an organization of political authority that is politically attractive to Americans–and self-reinforcing. Turning away from litigation may require re-legitimating and re-empowering other forms of authority." hypertext.niskanencenter.org/p/abundance-...

Very cool to see that @kevintobia.bsky.social's and my article got @lsolum.bsky.social's "Download it while it's hot!" recommendation on the Legal Theory Blog!

People often treat AI-generated legal texts as an issue of legal ethics (e.g., hallucinated citations) or industry economics (will AI replace associates?). But as legal institutions start receiving, processing, and using generative AI, it's going to affect all of us 🧵

I've got a new paper up with the inimitable @kevintobia.bsky.social: "Generated Legal Texts"—about texts generated by AI and used in legal institutions. These texts are arising frequently in legal contexts around the world, perhaps faster than many realize. And, we argue ...

A nice new opportunity for an AI+Law "Shark Tank" pitch session to get feedback on early stage paper ideas in time for the summer writing push, organized by @kevintfrazier.bsky.social at UT Law. Would definitely encourage anyone with a paper idea to check it out! docs.google.com/forms/d/e/1F...

In March, HUD declared it was “revising” (read: gutting) its Affirmatively Furthering Fair Housing regs. That was illegal. I got mad and submitted a comment. www.regulations.gov/comment/HUD-... It’s not in the 100 worst things happening right now, but it’s still MY thing. So, thread:

No. One: It’s a fact. Two: The judiciary is political. Three: The appointing president is often relevant to a full understand, hence newsworthy. Three (b): It is often these days a reminder that judges appointed by presidents of both parties are not acceding to Trump’s lawlessness.

I love this analogy

I started a blog! There is a lot of interesting stuff going on in the worlds of AI, consumer protection, and litigation that develops too quickly, or is too inchoate, for law review articles. I’ll be posting here occasionally—thoughts welcome:

This blog post also contains my favorite AI+Law anecdote so far from the teaching-and-talks circuit.

I enjoyed two recent empirical papers on AI+Law, but realized they have somewhat different implications—in particular, about whether AI tools are actually useful for lawyers, or are too likely to err. I wrote up a blog post thinking through how to reconcile them: www.wilftownsend.net/p/is-ai-actu...

Notably, from a law professor perspective, this includes accurately identifying a "dog not barking" issue on a long, issue-spotter-exam-style question that explicitly discussed many things but left an appellate jurisdiction issue completely unmentioned. No model had spotted that before.

I haven't spent much time with the new o3 model from OpenAI, but it is the first model to get all of the questions right on the informal testing of legal questions that I've been doing for a while when new models come out.

Several good things about this order. A small, technical detail that is good but easy to miss is the court's ruling on FRCP 65(c). In other times this would be unremarkable (or just absent from the order), but the Trump admin has tried to use Rule 65 to make this kind of civil action harder.

JUST IN: Trump-appointed judge in Texas blocks removals under the Alien Enemies Act, citing the SCOTUS ruling and the potential for a mistake as revealed in the Abrego Garcia case. This follows a similar order from NY judge on other AEA targets. storage.courtlistener.com/recap/gov.us...

NEWS: Jenner & Block sued to challenge Trump's executive order in D.C. federal court this morning. They are represented by lawyers from Cooley LLP, and are seeking similar relief to the Perkins Coie challenge. Complaint: storage.courtlistener.com/recap/gov.us...

A mandatory arbitration agreement that requires subjecting disputes to an AI system. What could go wrong?

This is a totally brilliant idea, and very well executed. More reviews of reviews!

Because Ezra Klein and Derek Thompson's "Abundance" is less a book than a discourse-generating machine (by design), I'm going to review the *reviews* (I also have a review of the actual book coming, I'm not a total weirdo). Roughly from worst to best:

In the growing world of AI litigation, class actions are going to be particularly important. In a new paper forthcoming in the Wash. U. law review, I look at how class actions will influence the efficacy of AI regulations, and how they might even open up some new options for regulators. 🧵

I am hiring a summer law fellow for my group! If you're pursuing a law degree and want to spend some time on legal/policy writing this summer—with the goal of an academic publication, like a law review article—please reach out! forms.gle/izNtBtEwiZaU...

Here's a weird thing: AI-generated articles are now quoting me for things I haven't said. For example, this piece provides quotes that I allegedly gave to Reuters. I haven't talked to Reuters about this. The quotes are *close* to what I might say, but not correct. opentools.ai/news/elon-mu...

A host of plaintiffs, including the NTEU, NCLC, NAACP, and an individual consumer, have sued the CFPB and Russell Vought challenging Vought's attempt to shut down the Bureau. Complaint filed by Gupta Wessler, @publiccitizen.bsky.social, and NTEU counsel: www.guptawessler.com/wp-content/u...

New day, new lawsuit against the Trump admin: latest challenging the halt of USAID funds, brought by plaintiffs incl. private contractors, small business group, nonprofits and the American Bar Association (gets $ for int'l rule of law programs) assets.bwbx.io/documents/us...

If you’re working on civil procedure projects, boy do we have a workshop for you! The tenth (!) annual Civil Procedure Workshop will be held at Richmond on June 2-3rd. Check out the link below and submit your abstract by March 3rd.

This is an optimistic take for what strikes me as a sticky problem. Not sure that legal academia counts as “science,” but I think we face similar questions about how to accurately value the production of scholarship, and those seem likely to get even harder.

Really important new lawsuit by @democracyforward.bsky.social and @publiccitizen.bsky.social , which are out there fighting the good fight.

This is an absolutely remarkable memo — and a betrayal of both the Constitution and the American political tradition. In the US, by *explicit contrast* with interwar fascist governments, the state is not supposed to be an extension of the personality of the chief executive.

second, courts are slow and procedural. Of course. But Trump is playing a different ground game altogether. I said this yesterday but the courts weren’t really “built” to withstand this kind of deluge of constitutional slop.

A nice profile of my colleague @markjia.bsky.social, whose scholarship always strikes me as a great balance between timeless issues and extremely relevant ones.

Nice to see a leaderboard/benchmark aimed specifically at hallucinations, in a context (retrieving information from documents) that seems like a common use case in law. I’ll be interested to see people’s assessments of this and whether it gets used much. deepmind.google/discover/blo...