Daniel Oliveira Zacarias
@indeeddaniel.bsky.social
Reposted by Daniel Oliveira Zacarias
Large Language models pose an existential threat to online survey research www.pnas.org/doi/10.1073/...
An AI can now evade a comprehensive suite of data quality checks, achieving a 99.8% pass rate on 6,000 trials of standard attention checks.
An AI can now evade a comprehensive suite of data quality checks, achieving a 99.8% pass rate on 6,000 trials of standard attention checks.
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
www.pnas.org
January 9, 2026 at 4:21 PM
Large Language models pose an existential threat to online survey research www.pnas.org/doi/10.1073/...
An AI can now evade a comprehensive suite of data quality checks, achieving a 99.8% pass rate on 6,000 trials of standard attention checks.
An AI can now evade a comprehensive suite of data quality checks, achieving a 99.8% pass rate on 6,000 trials of standard attention checks.
Reposted by Daniel Oliveira Zacarias
Round 1 results of our predicting replicability challenge are released!
The Challenge capitalizes on interest in using AI methods to do scientific tasks. Ten teams tried to predict replication outcomes from unreleased SCORE findings.
Results? Not so hot!
Post: www.cos.io/blog/predict...
The Challenge capitalizes on interest in using AI methods to do scientific tasks. Ten teams tried to predict replication outcomes from unreleased SCORE findings.
Results? Not so hot!
Post: www.cos.io/blog/predict...
November 24, 2025 at 5:30 PM
Round 1 results of our predicting replicability challenge are released!
The Challenge capitalizes on interest in using AI methods to do scientific tasks. Ten teams tried to predict replication outcomes from unreleased SCORE findings.
Results? Not so hot!
Post: www.cos.io/blog/predict...
The Challenge capitalizes on interest in using AI methods to do scientific tasks. Ten teams tried to predict replication outcomes from unreleased SCORE findings.
Results? Not so hot!
Post: www.cos.io/blog/predict...
Reposted by Daniel Oliveira Zacarias
Possibly weirded out my long-suffering colleagues with this idiosyncratic observation last week, and feel like it belongs in a meme.
#rstats #python #ripdavidlynch
#rstats #python #ripdavidlynch
October 24, 2025 at 12:07 PM
Possibly weirded out my long-suffering colleagues with this idiosyncratic observation last week, and feel like it belongs in a meme.
#rstats #python #ripdavidlynch
#rstats #python #ripdavidlynch
Reposted by Daniel Oliveira Zacarias
Against my better instincts, I have written some notes on how human probability judgements work and what you should expect from surveys that ask people to guess what proportion of the population is transgender. I hope never to speak of this matter again
Some notes on probability judgement – Notes from a data witch
For the love of fuck, literally nobody thinks that 20% of the population is transgender. Please stop sharing that ridiculous YouGov statistic
blog.djnavarro.net
September 21, 2025 at 3:38 PM
Against my better instincts, I have written some notes on how human probability judgements work and what you should expect from surveys that ask people to guess what proportion of the population is transgender. I hope never to speak of this matter again
F by far
For the rest of your life, you can only eat food from one of these regions. Which do you choose?
For more maps: brilliantmaps.beehiiv.com/s...
For more maps: brilliantmaps.beehiiv.com/s...
September 17, 2025 at 2:57 AM
F by far
Reposted by Daniel Oliveira Zacarias
Article looks interesting. Scientists need to put more energy toward saving the planet, biosphere, humanity.
The right, of course, will scream "Communism!!!"
www.nature.com/articles/s41...
The right, of course, will scream "Communism!!!"
www.nature.com/articles/s41...
August 21, 2025 at 2:51 PM
Article looks interesting. Scientists need to put more energy toward saving the planet, biosphere, humanity.
The right, of course, will scream "Communism!!!"
www.nature.com/articles/s41...
The right, of course, will scream "Communism!!!"
www.nature.com/articles/s41...
Reposted by Daniel Oliveira Zacarias
🚨 Acabamos de aprovar, na Câmara dos Deputados, a urgência do PL 2628/22.
Conhecido como “ECA Digital”, o Projeto protege crianças e adolescentes da exploração sexual nas redes sociais e freia parte dos abusos das big techs
Com isso, o PL poderá ser votado direto no Plenário da Câmara.
Conhecido como “ECA Digital”, o Projeto protege crianças e adolescentes da exploração sexual nas redes sociais e freia parte dos abusos das big techs
Com isso, o PL poderá ser votado direto no Plenário da Câmara.
August 19, 2025 at 10:48 PM
🚨 Acabamos de aprovar, na Câmara dos Deputados, a urgência do PL 2628/22.
Conhecido como “ECA Digital”, o Projeto protege crianças e adolescentes da exploração sexual nas redes sociais e freia parte dos abusos das big techs
Com isso, o PL poderá ser votado direto no Plenário da Câmara.
Conhecido como “ECA Digital”, o Projeto protege crianças e adolescentes da exploração sexual nas redes sociais e freia parte dos abusos das big techs
Com isso, o PL poderá ser votado direto no Plenário da Câmara.
Reposted by Daniel Oliveira Zacarias
Check out our new pre-print in which we argue something that might be obvious but doesn't seem to be obvious to everyone: Large Language Models (such as ChatGPT) do not simulate human psychology.
Large Language models (LLMs) do not simulate human psychology. That's the title of our new paper, available as preprint today (1/12):
arxiv.org/abs/2508.06950
arxiv.org/abs/2508.06950
Large Language Models Do Not Simulate Human Psychology
Large Language Models (LLMs),such as ChatGPT, are increasingly used in research, ranging from simple writing assistance to complex data annotation tasks. Recently, some research has suggested that LLM...
arxiv.org
August 13, 2025 at 12:37 PM
Check out our new pre-print in which we argue something that might be obvious but doesn't seem to be obvious to everyone: Large Language Models (such as ChatGPT) do not simulate human psychology.
Reposted by Daniel Oliveira Zacarias
The goal of a PhD is not to learn some facts or read a few papers or learn a bunch of techniques. The goal of a PhD is to learn independence, problem solving, how to finish things you start, resilience, & gain the ability to adapt & think creatively. Learning these things is hard.
August 13, 2025 at 5:28 PM
The goal of a PhD is not to learn some facts or read a few papers or learn a bunch of techniques. The goal of a PhD is to learn independence, problem solving, how to finish things you start, resilience, & gain the ability to adapt & think creatively. Learning these things is hard.
Reposted by Daniel Oliveira Zacarias
August 14, 2025 at 2:52 AM
Reposted by Daniel Oliveira Zacarias
Our (@malte.the100.ci, @ianhussey.mmmdata.io) commentary "cognitive dissonance in large language models is neither cognitive nor dissonant" has just been accepted for publication at PNAS!
Check out the thread below for a summary of our arguments.
osf.io/preprints/ps...
Check out the thread below for a summary of our arguments.
osf.io/preprints/ps...
August 1, 2025 at 4:51 PM
Our (@malte.the100.ci, @ianhussey.mmmdata.io) commentary "cognitive dissonance in large language models is neither cognitive nor dissonant" has just been accepted for publication at PNAS!
Check out the thread below for a summary of our arguments.
osf.io/preprints/ps...
Check out the thread below for a summary of our arguments.
osf.io/preprints/ps...
Reposted by Daniel Oliveira Zacarias
🎉New paper out @Psychological Review on the neuroscience of intergroup contact🎉
Led by the fantastic @margaretrwelte.bsky.social and Jas Cloutier!
@apajournals.bsky.social @spspnews.bsky.social @sansmeeting.bsky.social #PsychSciSky #SocialPsyc #AcademicSky
psycnet.apa.org/record/2026-...
Led by the fantastic @margaretrwelte.bsky.social and Jas Cloutier!
@apajournals.bsky.social @spspnews.bsky.social @sansmeeting.bsky.social #PsychSciSky #SocialPsyc #AcademicSky
psycnet.apa.org/record/2026-...
July 15, 2025 at 3:10 PM
🎉New paper out @Psychological Review on the neuroscience of intergroup contact🎉
Led by the fantastic @margaretrwelte.bsky.social and Jas Cloutier!
@apajournals.bsky.social @spspnews.bsky.social @sansmeeting.bsky.social #PsychSciSky #SocialPsyc #AcademicSky
psycnet.apa.org/record/2026-...
Led by the fantastic @margaretrwelte.bsky.social and Jas Cloutier!
@apajournals.bsky.social @spspnews.bsky.social @sansmeeting.bsky.social #PsychSciSky #SocialPsyc #AcademicSky
psycnet.apa.org/record/2026-...
Reposted by Daniel Oliveira Zacarias
The fucking head of fucking XBox Graphics used a fucking "AI" slop image to hire new engineers but couldn't tell that the fucking screen on the fucking monitor is facing backwards.
July 13, 2025 at 2:15 AM
The fucking head of fucking XBox Graphics used a fucking "AI" slop image to hire new engineers but couldn't tell that the fucking screen on the fucking monitor is facing backwards.
Reposted by Daniel Oliveira Zacarias
Attention LLM numpties
June 28, 2025 at 10:38 PM
Attention LLM numpties
Reposted by Daniel Oliveira Zacarias
Reposted by Daniel Oliveira Zacarias
America is one of those countries where if you host a global sporting event, inclusive messages have to be censored to so as not to anger the central government. In this case the message is “racism is bad” www.nytimes.com/athletic/643...
FIFA drops anti-racism, anti-discrimination messages for Club World Cup
www.nytimes.com
June 17, 2025 at 10:46 AM
America is one of those countries where if you host a global sporting event, inclusive messages have to be censored to so as not to anger the central government. In this case the message is “racism is bad” www.nytimes.com/athletic/643...
Reposted by Daniel Oliveira Zacarias
I've done a fair bit of research & read more on how to increase retirement savings. It's made me pretty pessimistic about every strategy except mandatory, universal pensions for baseline retirement income security. Lots of people just won't choose to save but will need $ later.
Just published in @jpube.bsky.social:
"New evidence on the efficacy of state-based retirement programs: The case of OregonSaves"
By John Chalmers, Olivia S. Mitchell (@mitchelos.bsky.social), Jonathan Reuter, & Mingli Zhong
www.sciencedirect.com/science/arti...
#econsky
"New evidence on the efficacy of state-based retirement programs: The case of OregonSaves"
By John Chalmers, Olivia S. Mitchell (@mitchelos.bsky.social), Jonathan Reuter, & Mingli Zhong
www.sciencedirect.com/science/arti...
#econsky
June 8, 2025 at 12:04 PM
I've done a fair bit of research & read more on how to increase retirement savings. It's made me pretty pessimistic about every strategy except mandatory, universal pensions for baseline retirement income security. Lots of people just won't choose to save but will need $ later.
Reposted by Daniel Oliveira Zacarias
One of the strongest illusions I’ve seen recently.
June 2, 2025 at 1:28 PM
One of the strongest illusions I’ve seen recently.
Reposted by Daniel Oliveira Zacarias
This is what a Cold Air Front looks like from Space
May 29, 2025 at 11:40 PM
This is what a Cold Air Front looks like from Space
Reposted by Daniel Oliveira Zacarias
Fitting a generalized mixed model with a gamma distribution log link and random slopes to reaction time data to arrive at precisely the same point estimate as the authors did by simply averaging and conducting a t-test:
May 28, 2025 at 5:22 PM
Fitting a generalized mixed model with a gamma distribution log link and random slopes to reaction time data to arrive at precisely the same point estimate as the authors did by simply averaging and conducting a t-test:
Reposted by Daniel Oliveira Zacarias
I generally recommend people to compare themselves to their past selves (i.e., have mastery goals). But if you *must* compare yourself to others, do it well and enter resources, social support, and random luck as covariates in your model.
May 23, 2025 at 5:16 PM
I generally recommend people to compare themselves to their past selves (i.e., have mastery goals). But if you *must* compare yourself to others, do it well and enter resources, social support, and random luck as covariates in your model.
Reposted by Daniel Oliveira Zacarias
“Next week will be quieter“, he mutters to himself for the 792nd week in a row
May 23, 2025 at 12:22 PM
“Next week will be quieter“, he mutters to himself for the 792nd week in a row
Reposted by Daniel Oliveira Zacarias
Bom link para jogar no grupo da Família:
youtu.be/Nn7UN7eeSXE?...
youtu.be/Nn7UN7eeSXE?...
Qualquer um está suscetível ao vício em bets | BdF Entrevista Altay de Souza
YouTube video by Brasil de Fato
youtu.be
May 23, 2025 at 1:53 AM
Bom link para jogar no grupo da Família:
youtu.be/Nn7UN7eeSXE?...
youtu.be/Nn7UN7eeSXE?...