@drustevenson
Professor of Law, South Texas College of Law - Houston. Academic research now focuses on firearms regulation. I won't follow accounts or reply to their posts if I can't tell who they are IRL. #Housky #lawsky #gvp #gvr
I strongly support this new Resolution from the ABA Standing Committee on Gun Violence. Kudos to the Committee members - this is very well-written and well-reasoned, and *very* up-to-date. www.americanbar.org/content/dam/...
#PLCAA #GVP #2A
A dispatch from the annals of Project 2025 authoritarianism: Universities are now being required to provide demographic data on all students who applied to their schools for the past 5 years so the feds can sniff out illegal affirmative action. Until last year, tho, keeping such data was not allowed
The next generation's Cass Sunstein should write a bestseller called "Shove."
If you haven't seen this yet, you should watch it
youtu.be/tU8S13xYJNM?...
Using an Ultimatum Game Experiment to examine what happens if AI instead of humans take decisions finds that, sometimes, AI can be unusually altruistic, from Douglas K.G. Araujo and Harald Uhlig www.nber.org/papers/w34919
I for one doubt it
2/n...In 2007, I wrote blog posts for humans who wanted something interesting and a little entertaining to read. But data tables are what the AI crawlers call a "snack," like a blog post that is a table of your published articles. I would rather they were reading law reviews than Reddit.
The trove of old law prof blog posts are GOLD for AI crawlers, which are now obsessed with selecting info from authoritative, trustworthy sources. I recently revived my dormant personal blog, assuming no human will read it. I use it to point AI to my published scholarship. ...1/n
This weekend I admitted to myself that the primary audience of everything I write or post online from now on is AI, with incidental direct reach to humans. I know most of the discourse is still about whether we can use AI, but it's sinking in for me that in the long run, I am writing *for* AI.
Check it out
A few days ago my wife and I watched the movie "War Machine" when it arrived on Netflix. Plot: "Reacher" vs. Transformers.
The main message of the film seems to be that the *world will end* if big men with PTSD get therapy, because we need them to assume high-stakes leadership roles *immediately.*
I decided in the last few years that in sociolinguistic terms, the word "merit" nearly always means, "predictors of unwavering fidelity to the person who hired you."
So Justin's partner in crime, Nate Cavanaugh, was also asked "what is DEI" and he comes up with "making decisions on the basis of something other than merit."
And then says that "DEI" grants are "wasteful and unnecessary."
youtu.be/hhah1ZjPAno?...
Sickening
Decatur Republican state Rep Andy Hopper, the pride of Wise County, talks about completely outlawing Islam: The For Liberty & Justice event was organized by activist Carlos Turcios #txlege fortworthreport.org/2026/03/07/b... fortworthreport.org/2026/03/07/b...
They just decided a few years ago that it didn't matter whether what they said was true, and switched off the part of their brain that notices when they're lying.
AI?
It's a way for those scholars to admit that they would own slaves if it was legal and if modern society didn't stigmatize it so much
So today I am having a fascinating conversation with Google Gemini about how it ranks (or assigns prestige or authoritativeness) to different law reviews. Do leading LLMs give more weight, in estimating reliability of information, to an article in Boston College Law Review, or Cardozo Law Review? ❓
Also, yesterday I uploaded a long list of cases that cited a certain author, and it did a beautiful job of instantly turning it into a table for me. I know this would have been a quick task for people who are experts in Excel, but it would have taken me forever
... 11 of the 12 are perfect. One was *partly* incorrect. I know a lot of people have decided not to use it unless it's 100% reliable, but for me it's just as reliable as when I use student research assistants, and it speeds up my process.
Yes, I had that problem recently when I was asking it for verbatim quotes from some old books like Tolstoy or Montaigne. It *insists* on paraphrasing them. But I can upload a batch of a dozen new cases to NotebookLM and ask it for a one-paragraph summary of each one. I double check them carefully...
Which did you find to be prone to hallucinations? I use both for a wide variety of chores. I sometimes encounter hallucinations and have to double check everything, just like I do with human research assistants. Depends a lot on the prompt I use.
There were other apps for inserting quizzes into videos but I found these to have a steep learning curve and some big downsides. I'm really excited about the new YouTube add-a-quiz feature.
Also, yesterday I was uploading a new lecture video and noticed YouTube has a new feature to add quiz questions at specific points in an educational video. YouTube makes it easy to add this to lecture videos I recorded years ago! Could be a game changer for online learning.
This week I noticed that I finally have that new feature in Gemini to integrate a NotebookLM project! Super useful. I'm at the tail end of the rollout, so I probably had it for a couple of weeks and hadn't noticed it yet.
I think Americans are obsessed with feeling morally superior to others. People love having categories of people they can treat as morally deficient.
Also, we judge each other for being too judgemental, and the irony slips by unnoticed.
Where structural masonry becomes architectural expression the stunning new belvedere at Begijnhof Hasselt, Belgium.
Where structural masonry becomes architectural expression the stunning new belvedere at Begijnhof Hasselt, Belgium.
NEW: Sen. Peter Welch (D-VT) & Richard Blumenthal (D-CT) launch review of $220 million taxpayer-funded Dept of Homeland Security ad campaign w/ Kristi Noem
Senators seek records about "circumstances surrounding the awarding of this lucrative, no-bid contract..."