There's growing evidence that LLMs can p-hack.

But p-hacking also points to something bigger: a data science multiverse of defensible analytical choices.
We wrote a paper (arxiv.org/abs/2602.18710) on using LLM agents to map this multiverse systematically. 🧵

California needs to pass zoning reform — and build new housing — in order to prosper OPINION: “If we want to revitalize our cities, we need to change how we build. That’s exactly what California Senate Bill 79 sets out to do,” Tom Steyer writes.

Tom Steyer lays out the facts on why we need real state level action on zoning reform: “Here’s the reality: Cities like San Francisco and Los Angeles don’t lack land. What they lack is permission to build the housing people desperately need.”

zohran beating cuomo

would love to get philosophers engaging with aspects of the debate I find most interesting here—esp those that I think economists tend to overlook. will try to write about it at some point (low on my priority list as a computer scientist but high on it as a political junkie living in NY)

oh yeah that’s essentially what I meant (i.e that it’s dealing with a super weak argument, not necessarily an artificially constructed / misrepresented one) my b

of course, I’m sure you did and are

but that IS what the debate is about. this thread is dealing with a strawman

not to be taken lightly, podcasters are some of the most powerful people in the world right now

very relatable totally understood 👍

these models always start simple so that you can prove nice theorems about them, they’re not meant to represent the real world but they do usually intend to build some intuition. they are also useful independently of whether they represent reality or not (like thought experiments)

I say this too often but I think it’s relevant here: every philosopher, economist and computer scientist should read Amartya Sen, that might get us all speaking the same language on ethics

interesting! can you elaborate on / point out what you think is wrong there?

tbf none of those examples seem to appear in actual AI/ML conferences

It could take another decade or two to get at the reality here. In the meantime we will certainly need to look skeptically at EdTech schemes and cheap shortcuts. But overall predictions of doom are not warranted yet.

critical work on assumptions underlying causal inference as practiced in social sciences is also a rich area e.g. see www.cambridge.org/core/journal...

not sure if this fits what you’re looking for but in CS I regularly see critical work on assumptions in various forms e.g. around notions of security, privacy and fairness. here’s one relevant survey: www.annualreviews.org/content/jour...

Opinion | West Point Is Supposed to Educate, Not Indoctrinate

www.nytimes.com/2025/05/08/o...

Graham Parsons, professor of philosophy at West Point, announces his resignation in the face of new policies "eliminating courses, modifying syllabuses and censoring arguments to comport with the ideological tastes of the Trump administration."

The Sources of Researcher Variation in Economics Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

"146 research teams each completed the same causal inference task three times each... We find that even when analyzing the same data, teams reach different conclusions."

Everything matters, from data cleaning to functional form choices!

www.nber.org/papers/w3372...

"Science is an investment.

We will put forward a new 500 million package for 2025-2027 to support the best and the brightest researchers and scientists from Europe and around the world."

— President @vonderleyen.ec.europa.eu at the ‘Choose Europe for Science' event at La Sorbonne 🇫🇷

Opinion | I Teach Computer Science, and That Is All Politics has no place at universities or in the classroom.

I mention Computer Science in particular because the Times happens to have an opinion piece by a Harvard CS prof who argues that there should be a firewall between the "academic and the political." He seems to think that ideology has nothing to do with his field. www.nytimes.com/2025/05/02/o...

What does it mean to be an author when using AI to produce a text?

"Distant Writing: Literary Production in the Age of Artificial Intelligence"
new paper freely available at SSRN
papers.ssrn.com/sol3...

Pete Hegseth’s Patriotic Duty Is to Resign His incompetence is putting America’s security at risk.

Hegseth has been a proven failure as a secretary of defense. If the president still refuses to fire him, he should resign.

www.theatlantic.com/ideas/archiv...

Here we go! @kateconger.com was verified - not by Bluesky but by the NYTimes, one of the "trusted verifiers"

Gonna keep saying it: a healthy digital society should distribute power!

How Universities Became So Dependent on the Federal Government For decades, universities got billions in federal dollars for research. The relationship was mutually beneficial, until President Trump decided it wasn’t.

This framing is nuts, and the assumption is that public funding for a public good is somehow untoward and illicit. “The schools took the money” and became “warily beholden to the whims of politicians in Washington.”/1
www.nytimes.com/2025/04/18/u...

Observe that the one that complied and the one that didn't comply are treated exactly the same way.

We Must Save Public Media to Change It We need public media more than ever—it’s too precious to let Trump defund it. But to live up to its democratic promise, we must support public media to serve our needs.

New piece in The Nation arguing that we need public media more now than ever. We must defend what we have with an eye toward expanding and restructuring public media in the future.

philosophy/ethics is actually a growing area of interest within the CS research community and that’s a good thing

For the Trump Administration, it is a matter of principle that the President can disappear innocent people to foreign gulags without due process or judicial review.

surveillance tech company founded in 1984. perfection

good quote