@suffolklitlab.org
The Legal Innovation & Technology Lab is part of the clinics at Suffolk University Law School. We build things that help folks navigate and understand legal systems. Learn more at: https://suffolklitlab.org/ Law+Tech bot feed @news.bot.suffolklitlab.org
I knew working on suicide research was going to be heavy.
But what's actually starting to get to me is reading all of these chatbot suicide laws and legislative proposals that call for measures that have been shown to exacerbate crisis.
www.governor.ny.gov/sites/defaul...
The very existence of commercially available glasses with Internet-connected cameras sold by Meta, a company built entirely on hiding the costs of services in negative privacy externalities, is an indictment of U.S. tech law and policy. This is an unfixable product from a fundamentally bad company.
OpenAI hit with lawsuit claiming ChatGPT acted as an unlicensed lawyer www.reuters.com/legal/legali...
A new lawsuit alleges Google’s chatbot sent a Florida man on missions to find an android body it could inhabit. When that failed, it set a suicide countdown clock for him. www.wsj.com/tech/ai/gemi...
When shaping your research agenda, your objective is to find the weirdest niche possible that still has the potential to change everything.
This is fucked up and also an interesting question about right of publicity.
So @davidcrespo.bsky.social made me blog: lu.is/2026/03/on-m...
TLDR: generalist LLMs are not lawyers, and evaluating them that way is a waste of time. Evaluating LLMs with useful specialized prompts (and eventually, with specialized legal harnesses) is where the work must happen.
👀
This is insane. I suspect it won't come to this, but I'd love to see the DoD try to invoke the DPA and Anthropic sue arguing that as software this was compelled speech. It would make such a great case for my AI & the Law course. Either way it's a great hypo and super troubling that DoD wants this.
Join us in Boston (or online) on April 13, 2026 for our annual Legal Innovation & Technology conference: #LITCon2026. IRL attendees get breakfast ☕️ 🥐, a hot lunch 🍗 🥗, and a late-afternoon reception with hors d'oeuvres and drinks 🍻 🍷 🧀.
Register here: suffolklitlab.org/events/lit-c...
🚨 Call for rapid-fire talks! 🚨 Give us your best pitch for a 5-min talk on legal innovation and technology for a chance to present at #LITCon2026 on April 13th in Boston! suffolklitlab.org/events/lit-c...
“Generative AI breaks long-standing correlations that society uses to infer things like effort, sincerity, authenticity, and credibility. Once these signals erode, we don’t automatically get something better.”
jessicahullman.substack.com/p/zeynep-tuf...
THIS.
I am sad that in this edition of AI Bootcamp for lawyers “computational thinking” was not going to fit in the syllabus, but thinking in “if this then that” is more powerful than ever.
First, I heard a rumor that Westlaw turned down the opportunity to work on this with UC SF.
Second, our APIs are now powering seventeen foot high art in San Francisco!!
A lesson about networks that I’m kind of proud of: at the beginning of the year, I gave my 200 students a dumb survey: fav books, foods, etc. I told them they could use pseudonyms and that I’d share their responses with the class. We’ve been using that data in various ways: to make points about +
If for some reason you want a glimpse into the current conversations of a bunch of law-type folks, #SCOTUS decision perhaps, @icymilaw.org has you covered. Behold their law list.
“Here, I thought, are the two strands of robotics: one useful and invisible, the other theatrical and redundant.”
harpers.org/archive/2025...
Lab co-director @qsteenhuis.bsky.social makes the case that it is morally wrong to withhold carefully tested AI tools from unrepresented litigants when the alternative is ChatGPT or nothing at all.
The solution to this is banning advertising to teenagers, not banning teenagers from the internet.
DOI's are a great example of the critical infrastructure that librarians quietly enable.
Last night, someone on this here website told me that no one is using AI agents yet and they're still way far in the future. But it's already happening.
A friend last year told me that most people are distracted by the idea that AI is "chatbots." I didn't get that at the time, but now I think I do.
*Great* session on public involvement in science, governance, and resistance to AI today at the Participatory AI Research & Practice Symposium at the India AI Impact Summit.
You can see the full list of presentations here:
docs.google.com/document/d/1...
I fear my favorite exercise was lost in last week's mega thread. It asks students to confront the possibility of competing and mutually exclusive concepts of the good, and ask if policies that deliver on one measure might have to change when the context changes. You should check it out.
#lawlibrarians assemble!
"News Publishers Are Now Blocking The Internet Archive, And We May All Regret It."
www.techdirt.com/2026/02/13/n...
@masnick.com is right: "In our rush to punish #AI companies, we’re destroying public goods that serve everyone."
#Copyright #InternetArchive #Journalism #Publishers
@archive.org
After having its code request rejected, an AI agent allegedly published a hit piece on the (human) gatekeeper.
While the idea of autonomous defamation bots is intriguing (in a sci-fi kind of way), there are critical gaps in this story. 🧵
www.fastcompany.com/91492228/mat...
I think I've asked this on here before, but can't recall:
Has anyone seen/written a paper on how statistical significance is a burden of proof, and so the law's mindless focus on significance for admissibility effectively imposes the SAME burden on BOTH parties, in civil AND crim cases alike?
So this reminds me of something:
When law students ask me what to take in undergrad, my usual answer is "anything, really."
But probably a better answer would be "it might be helpful to grab a statistics class or two."
Last week I ask some law students the question below. Now it's your turn. Click through to the thread for a link to a simulation to help you clarify your thinking, then continue with the thread to see the correct answer.
This is so, so good. If you have units on algorithmic bias, or are just interested in incorporating simulations/interactive elements into your teaching, check this out.