←
Jon Green2,496 posts
Jon Green avatar

Jon Green

@_Jon_Green

Assistant professor @DukePoliSci jongreen.bsky.social

◷ Joined February 2012▧ 274 Photos and videos
Jon Green avatar
Jon Green@_Jon_Green·1 Jan 2018

Hello new followers! You probably came for politics tweets, and you’ll get them. But you’re also gonna get a lot of Columbus sports tweets and pictures of my dog.

Post attachmentPost attachment
Alexander McCoy avatar
⟳ Jon Green retweeted
Alexander McCoy@AlexanderMcCoy4·Nov 17

My repeated experience working on tech policy is that people who work in tech have VERY bad mental models for the political system/government, and are often operating on a shallow libertarian stereotype of it being a sclerotic behemoth primed for disruption. But it’s actually a high-signal, hyper-competitive (and thus high-efficiency) realm! If things seem slow or gridlocked, it’s because a complex network of sophisticated opponents are fighting each other to a draw, not because everybody is a stupid, ignorant and lazy bureaucrat.

I also think everyone who believes this is just wrong about how fast Congress can move if they conclude that there's a real emergency.

Jon Green avatar
Jon Green@_Jon_Green·Nov 13

I guess it's obvious why/how this happens but, man, imagine automating a job application to PANGRAM of all places

We ask this simple question in our job application, and it cuts down on our application review by 60%. In the last month, we've seen the following: - 27 applicants outright say they're an AI - 1,217 applicants name an ML concept - 1051 applicants say N/A or "Not specified in my profile" The remaining 2002 applicants name a snack. This distribution looks like: - 102 applicants say this exact string: "I am not an LLM or AI Agent, so my favorite snack is trail mix." - 63 applicants say "I am not an LLM or AI Agent, so my favorite snack is dark chocolate." - 20 applicants say "I am not an LLM or AI Agent, so my favorite snack is trail mix because it provides a good balance of energy and nutrients during long coding sessions." - 10 applicants say "I am not an LLM or AI Agent, so my favorite snack is trail mix because it provides a good balance of energy and nutrients during long work sessions." - 16 applicants say "I am not an LLM or AI agent, so my favorite snack is dark chocolate." - 47 applicants say "My favorite snack is trail mix." - 16 applicants say "My favorite snack is dark chocolate." - 143 applicants say "Almonds" or "almonds" - 108 applicants say "Dark chocolate" or "dark chocolate" - 23 applicants say "Dark chocolate almonds" and then we have a long tail of ~1500 interesting or unique snacks, which we can use to keep the office pantry stocked when a new hire joins

Post attachment
Kyle Saunders avatar
⟳ Jon Green retweeted
Kyle Saunders@profgoose·Nov 12

1/ Ask an AI about a conspiracy theory and it'll almost always tell you the claim is false. Whether it also tells you your concern is reasonable depends on whether the theory blames your side or the other one. 135,845 conversations, seven @OpenAI models. osf.io/preprints/psyarxiv/4y…

Post attachment
Jon Green avatar
Jon Green@_Jon_Green·Nov 12

unironically if this is all just downstream of which things you think are important, one of these orgs could simply vibe code a dashboard where *you* pick how much weight to give to which criteria and have it spit out a ranking of your choice

City Journal's rankings of colleges includes factors like "Do you have a lot of ROTC kids?" and "Do you have lots of religious organizations". They also give bonus points if you join random orgs their donors support like "Heterodox Academy" or "BridgeUSA" which seems grifty.

Jon Green avatar
Jon Green@_Jon_Green·Nov 11

we also asked about recognition of/favorability toward the Christian flag. not the main point, but relevant as flags are pretty important for the idea of nationhood!

Post attachmentPost attachment
Jon Green avatar
Jon Green@_Jon_Green·Nov 11

importantly, we also show that you will completely miss this if all you do is estimate a dimension and use scores on that dimension in a regression. observing a relationship between attitudes and outcomes tells you ~nothing about the prevalence or extremity of the ideology

Post attachment
Jon Green avatar
Jon Green@_Jon_Green·Nov 11

in these new data: - responses look like meaningful attitudes! but: - ~everyone fails to check at least one box we'd expect an ideologue to check - those who ID as CN'ists are ambivalent on the worldview - those most willing and able to define CN are those who hate it

Jon Green avatar
Jon Green@_Jon_Green·Nov 11

so we ran our own survey, which has: - a worldview battery spanning three theoretical components of the ideology, each component including variation in item direction and difficulty - an identity measure - an open-ended "what does CN mean?" item

Jon Green avatar
Jon Green@_Jon_Green·Nov 11

the earlier version used publicly available data. for our purposes, these had: - too few items - not enough variation in item difficulty - not enough variation in item direction (whether agreement is pro- or anti-CN) - unclear mapping to theoretical components of the ideology

Jon Green avatar
Jon Green@_Jon_Green·Nov 11

big update to this working paper: new data, new (simpler!) methods, similar story: the empirical literature on Christian nationalism has measured something interesting, but probably not the niche ideology that it's worried about at the elite level osf.io/wbdut/files/vbky9

Post attachmentPost attachmentPost attachmentPost attachment
David A. Hopkins avatar
⟳ Jon Green retweeted
David A. Hopkins@DaveAHopkins·Nov 10

If you're wondering whatever happened to the moderate bloc of House Republicans, the bluest Republican-held district in the country voted for Trump by 9 points in 2024 and will be represented by Paul LePage:

Reminder: when a majority party in the House gets crushed in an election, the resulting caucus/conference becomes more extreme, not less. To wit:

Jon Green avatar
Jon Green@_Jon_Green·Nov 9

maybe! but the people who voted for Mangeris understood that the actual competition was between Boebert and Laubacher and decided not to participate in it. not obvious that taking Mangeris off the ballot forces them to participate, let alone vote for Boebert

Another election handed to Democrats because of the libertarians.

Post attachment
Jon Green avatar
Jon Green@_Jon_Green·Nov 7

either the voters share my preferences and are therefore redundant, or they don’t share my preferences and are therefore wrong

Human disempowerment now. It's time for the age of the folly and cruelty of human politics to end. The median AI is so far above the median human voter in both knowledge and decency it's embarrassing. Humanity should just take the L.