frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Dishonest People Self-Select into Public Service (In China) [pdf]

https://bfi.uchicago.edu/wp-content/uploads/2026/01/BFI_WP_2026-17.pdf
3•marojejian•1h ago

Comments

marojejian•1h ago
Found this research via the excellent episode of "Not Another Politics Podcast" (and adapted the title from them): https://podcasts.apple.com/us/podcast/do-dishonest-people-se...

I suggest listening to it.

Now of course, I am shocked, shocked that politicians lie a lot. But moving beyond the cliche, I think is evidence is fascinating and important.

This research follows on from a tradition which supports that dishonest people self select into public service, in countries where it is more corrupt (so the rewards seem larger): e.g. India: https://www.aeaweb.org/articles?id=10.1257/pol.20150029

And select out of of public service where it is trustworthy (e.g. Denmark): 2) https://www.aeaweb.org/articles?id=10.1257/pol.20170688

I don't think this is specific to China at all (not to say there isn't cultural variance). It just happens to be a place where we have good data.

This research adds great data, and interesting insights:

- Plagiarism 15 years ago correlated with dishonesty today, and for a small payout. This is consistent with the theory that that some people are predictably dishonest, and even when the rewards are small.

- People who plagiarize are more likely to be promoted in 5 years.

- In the subset of judges in the sample, they could detect that the judgements of plagiarists were more friendly to the high-status litigants, but only if the trial had not been live-streamed to the internet.

Spillover effects: - people who work for the "dishonest" Judges end up emulating their mentor's ruling patterns over time.

- "Dishonest" Lawyers secure higher win rates when they face "dishonest" judges.

The author concludes: Institutions are only as good as the people inside them.

My experience supports this. There are always natural incentives to cheat, and thus it behooves society to construct costs to combat this. I don't think our modern society (private or public) does this well, and thus we are proportionally led by those who exploit the system, because they succeed by selection.

austin-cheney•1h ago
Politicians lie because they are incentivized to do so. The simple solution is to increase the costs of ethical violations to an absurd level and keep politicians’ conduct under constant investigation.

In the US Army, for example, 25% of general officers are constantly under investigation. This does not mean general officers are constantly breaking the law but that conduct is continuously examined in case they do. The consequences are sufficient to deter so much of the self serving behavior we commonly associate with US politicians. At that point the cost becomes too high for narcissistic personalities to enter that line of work.

Narcissistic personality deterrence is what we want from politicians, which cannot happen when politicians largely police their own conduct hidden behind secret alliances.

Hardwiring LLM model weights to ASIC

https://www.eetimes.com/taalas-specializes-to-extremes-for-extraordinary-token-speed/
1•d_silin•1m ago•0 comments

From chickens to humans, animals think "bouba" sounds round

https://arstechnica.com/science/2026/02/newly-hatched-chickens-form-the-same-sound-association-we...
1•pseudolus•1m ago•0 comments

URL Reputation Checker (CLI and Web), Works Without API Keys

https://github.com/newuni/url-reputation
1•newuni•3m ago•2 comments

Running OpenClaw safely: identity, isolation, and runtime risk

https://www.microsoft.com/en-us/security/blog/2026/02/19/running-openclaw-safely-identity-isolati...
1•ChrisArchitect•7m ago•0 comments

Apollonius circles vs. the three-body problem

1•isaacbowen•8m ago•0 comments

Is AI helping you to do work, or just to do stuff?

https://coyotetracks.org/blog/ai-work-vs-stuff/
1•cdrnsf•11m ago•0 comments

Generate a 3D-printable STL terrain tile of the Lunar surface

https://moontoprint.com/
1•1970-01-01•16m ago•0 comments

Bluesky Head of Product soliciting feedback on upcoming search upgrade

https://bsky.app/profile/alexbenzer.com/post/3mfauv2c3gk2u
2•verdverm•16m ago•0 comments

Archive.today CAPTCHA page executes DDoS; Wikipedia considers banning site

https://arstechnica.com/tech-policy/2026/02/wikipedia-might-blacklist-archive-today-after-site-ma...
3•Computer0•16m ago•0 comments

The imminent risk of vibe coding

https://basta.substack.com/p/the-imminent-risk-of-vibe-coding
1•donutshop•16m ago•0 comments

Can language models synthesize scientific literature?

https://openscilm.allen.ai
1•bonsai_spool•17m ago•0 comments

Pi for Excel: AI sidebar add-in for Excel, powered by Pi

https://github.com/tmustier/pi-for-excel
1•rahimnathwani•18m ago•0 comments

Show HN: Three new models by KittenML. <25 MB Open-source TTS. Highly Expressive

https://kittenml.com
1•rohan_joshi•27m ago•2 comments

Ditching Discord

https://wiki.alopex.li/DitchingDiscord
2•pabs3•27m ago•0 comments

Rented Virtue

https://minutes.substack.com/p/rented-virtue
1•barry-cotter•28m ago•0 comments

The Embarrassing Truth About Tariffs

https://www.wsj.com/opinion/kevin-hassett-tariffs-study-federal-reserve-bank-of-new-york-donald-t...
3•petethomas•28m ago•0 comments

Show HN: What Chicken Said

https://archive.org/details/chicken_202104
1•HocusLocus•30m ago•0 comments

Open Source Software Projects Are Brands

https://reidkleckner.dev/posts/oss-projects-are-brands/
1•matt_d•30m ago•0 comments

New Startup Ideas (2011)

https://docs.google.com/spreadsheets/d/1pg7-ql_6IE8qx0JpnvoHtllUAVS83EFNMt8TM_MBZDU/edit?hl=en_GB...
1•nvr219•33m ago•0 comments

Unpaid – AI payment reminders that verify, escalate, and auto-reply

1•SnapExp•33m ago•1 comments

Anthropic legal request: OpenCode removes Claude subscription support

https://github.com/anomalyco/opencode/commit/973715f3da1839ef2eba62d4140fe7441d539411
3•DrammBA•34m ago•1 comments

Perfect heat rectification and circulation with nonreciprocal radiative surfaces

https://journals.aps.org/prresearch/abstract/10.1103/7p58-n6yv
1•PaulHoule•38m ago•0 comments

Rebuilding the American Dream, One Row House at a Time

https://www.thefp.com/p/rebuilding-the-american-dream-one
1•mhb•39m ago•0 comments

Why Traditional DLP Fails in the Age of LLMs

https://chris-s-lambert.com/why-traditional-dlp-fails-llms.html
1•clambert4•39m ago•0 comments

There Are Fewer Excuses

https://ctts.substack.com/p/there-are-fewer-excuses
1•threefiftyone96•42m ago•0 comments

NASA’s Artist’s Cooperation Program

https://airandspace.si.edu/explore/stories/nasa-art-program
1•animal_spirits•43m ago•0 comments

Ask HN: What is the future of open source?

1•andreybaskov•47m ago•0 comments

RaidenFTPD

https://www.raidenftpd.com/en/
2•metadat•47m ago•0 comments

Skills Manager

https://github.com/razbakov/skills-manager
1•razbakov•49m ago•0 comments

Lords of the Ring: The cultural politics of sumo wrestling

https://harpers.org/archive/2026/03/lords-of-the-ring-joshua-hunt-cultural-politics-sumo-wrestling/
1•zdw•50m ago•0 comments