Marcelo Alves

The Irrational Decision How the computer revolution shaped our conception of rationality—and why human problems require solutions rooted in human intuition, morality, and judgment

I’m excited to announce that my new book, _The Irrational Decision_, is now available for pre-order from Princeton University Press.

In a first, Google has released data on how much energy an AI prompt uses It’s the most transparent estimate yet from one of the big AI companies, and a long-awaited peek behind the curtain for researchers.

Ooh, look. When we covered the climate impacts of machine learning (“AI”) on Rare Earth in June we found it really frustrating that none of the tech companies would tell anyone the water and energy costs. Now Google has made at least some of it public. More of this! (and preferably raw data too).

More zero-click searches, same revenue for Google, less visibility for publishers?

Three things to read together on (1) how users respond to AI summaries, (2) what the company says it means for volume of queries and revenue, and (3) outside analysis on publishers’ visibility 1/4

Puzzled by all these AI policies?
This Global AI Governance Matrix is very useful 👇
(credits to Walter Pasquarelli!)

Data centers das big techs viraram símbolo de “progresso” no Brasil: ganham isenção, água, energia e terrenos. É a mesma história de sempre: exploram o território enquanto lucram lá fora. Um colonialismo versão 2.0, mas agora com cabos e servidores no lugar de navios.

a couple of hours before my keynote, I went through an intense negotiation with the organisers (for over a hour) where we went through my slides and had to remove anything that mentions 'Palestine' 'Israel' and replace 'genocide' with 'war crimes'

1/

A ideia de liberdade de expressão e o caso Musk X Moraes Café da Manhã · Episode

Pra quem quer entender o que Trump fala na carta sobre as big techs, vale ouvir o podcast da @folha.com em que comento o uso de “liberdade de expressão” e “censura” mencionadas por Trump no anúncio das tarifas open.spotify.com/episode/1ZV3...

McDonald’s AI Hiring Bot Exposed Millions of Applicants’ Data to Hackers Who Tried the Password ‘123456’ Basic security flaws left the personal info of tens of millions of McDonald’s job-seekers vulnerable on the “McHire” site built by AI software firm Paradox.ai.

McDonald's uses an AI bot called "Olivia" for hiring. A pair of hackers found they could access every conversation job applicants had with it—including all the personal info they shared—by exploiting security vulnerabilities as basic as using the password "123456". www.wired.com/story/mcdona...

There is no ethical AI. The tech is based on stolen intellectual property, anti-solidarity between workers, labor exploitation, environmental injustice, climate destabilization, structural racism, and gender inequalities.

The United States, that "beacon of democracy", has intervened in 128 elections since 1946 to shape political outcomes in their own interests.

Quantifying Narrative Similarity Across Languages - Hannah Waight, Solomon Messing, Anton Shirikov, Margaret E. Roberts, Jonathan Nagler, Jason Greenfield, Megan A. Brown, Kevin Aslett, Joshua A. Tuck... How can one understand the spread of ideas across text data? This is a key measurement problem in sociological inquiry, from the study of how interest groups sh...

I am thrilled to share a new article in Sociological Methods & Research, “Quantifying Narrative Similarity Across Languages”. My co-first author Sol Messing and our collaborators developed a new approach to measuring “narrative similarity” between texts: journals.sagepub.com/doi/10.1177/...

The last few months have been the most impactful by 404 Media yet. Stopping ICE data access; triggering changes at AI companies, more. We continue to be the most impactful tech outlet on the internet, and our subscribers let us do this work. Here are some of those recent examples:

Meta Invents New Way to Humiliate Users With Feed of People's Chats With AI In an industry full of grifters and companies hell-bent on making the internet worse, it is hard to think of a worse actor than Meta, or a worse product that the AI Discover feed.

Meta's AI discover feed is (still!) leaking user conversations with its AI. With its scale/import in shaping people's perception of reality, it is hard to think of a worse actor in the AI space than Meta, or to imagine a more invasive, user-hostile product than this

www.404media.co/meta-invents...

A guide to investigating digital ad libraries How to dig into ad libraries from Google, Meta and other platforms to track spending, narratives, campaigns and more

Ad libraries from Meta, Google and other platforms can be incredibly useful for reporting and investigations. Track campaigns and narratives, hunt for disinfo & scams, and more.

Here's the Indicator guide to using and investigating ad libraries:
indicator.media/p/a-... #OSINT

Google AI Overviews Says It's Still 2024 When asked to confirm the current year, Google’s AI-generated top result confidently answers, “No, it is not 2025.”

When asked to confirm the current year, Google’s AI-generated top result confidently answers, “No, it is not 2025.” www.wired.com/story/google...

Data centers em alta, mas a que custo? 🌿

O Brasil está vivenciando uma corrida por mais infraestrutura digital, com a rápida expansão dos data centers.

Entenda riscos e o que está sendo feito! ⬇️

Epic triple threat!!

A secret experiment that turned Redditors into guinea pigs was an ethical disaster—and could undermine research that’s only getting more urgent, @tombartlett.bsky.social reports. https://theatln.tc/xtKXNOyA

Cover Image for: Risky Research: An AoIR Guide to Researcher Protection and Safety AoIR Risky Research Working Group Contributors: Alice Marwick, Dafna Kaufman, Jacob Smith, Patricia Aufderheide, Jessica Beyer, Emma L. Briant, Coppélie Cocq, Laura Dilley, Sam DiBella, Radhika Gajjala, Kamile Grusauskaite, Alex D. Ketchum, Zelly Martin, Janice Metzger, Erin McInerney, Rachel Moran, John McNutt, Suay Melisa Oezkula, Victoria O’Meara, Riccardo Nanni, Carolina Parreiras, Katy Pearce, Ryan Payne, Meredith Pruden, Christian Sandvig, Caighlan Smith, Sam Srauy, Zeerak Talat, Leonie Tanczer, Robert Tynes, Antonia Vaughan, Shenja van der Graaf, Courtney Vowles, Michele White.

We are delighted to share the publication of Risky Research: An AoIR Guide to Researcher Protection and Safety, the culmination of over two years of collaborative effort by the AoIR Risky Research Working Group.

aoir.org/riskyresearc...

Anatomy of an AI Coup | TechPolicy.Press DOGE is gutting federal agencies to install AI across the government. Democracy is on the line, writes Tech Policy Press fellow Eryk Salvaggio.

"We are in the midst of a political coup that, if successful, would forever change the nature of US government. It is not taking place in the streets. There is no martial law. It is taking place cubicle by cubicle in federal agencies and in the mundane automation of bureaucracy." @eryk.bsky.social

The AI jobs crisis is here, now It's not coming, it has already arrived.

The AI jobs crisis is not coming—it's here right now. It just might not look the way many thought it would.

www.bloodinthemachine.com/p/the-ai-job...

Parabéns!

Google loses adtech monopoly case A loss for Google.

the US Department of Justice has won its antitrust case against Google, which accused the company of operating a monopoly in the ad tech industry. This is the 2nd time in a year that a federal court has ruled Google acted illegally to maintain a monopoly www.theverge.com/news/650665/...

Scammers Used OpenAI to Flood the Web with SEO Spam A new report from SentinelOne exposed the inner workings of AkiraBot, a program that bypassed CAPTCHAs and used AI-generated messages to target 420,000 websites.

Scammers used OpenAI as part of a bot that flooded the web with SEO spam. Also bypassed CAPTCHA www.404media.co/scammers-use...

Publicamos uma reportagem especial na BBC News Brasil, feita nos últimos quatro meses, que identificou mais de 200 contratos públicos de pavimentação firmados com construtoras autuadas ou que já compraram pedras extraídas de pedreiras que empregavam trabalhadores em situação análoga à de escravo.

Imperdível! Steven trabalha com maestria e acessibilidade o tema. Não menos importante, é uma abordagem crítica sobre o assunto, trazendo alternativas ao marasmo de algumas interpretações hegemônicas. 👇

📢 #Callforpapers: “From Fact-Checking to Community Notes: Structural Shifts in the Fight Against Disinformation on Social Media”

Editors:
Ramón Salaverría, University of Navarra, Spain
Lucas Graves, University of Wisconsin-Madison, U.S.
Raquel Recuero UFPEL/UFRGS

revistas.unav.edu/index.php/co...

During an inspiring three day workshop organized by @sahanaudupa.bsky.social, a group of expert researchers, practitioners and policy makers presented and discussed the results of their research about extreme speech on encrypted messaging platforms. 1/4

This information could be used to chill out and just let people try to make the best things they can without sweating imaginary numbers, or disregarded in favor of freaking out over what we imagine those numbers to mean. Let me just take a big sip of coffee while I wait to find out which way it goes

Anti AI piece in the New York Times and another in CNN from the excellent @amorrow.bsky.social. Something is changing.

amp.cnn.com/cnn/2025/03/...