Sign in

Daniela Tafani

@danielatafani.bsky.social
175 followers 147 following 157 posts
PostsRepliesMedia
Reposted by Daniela Tafani
Meg Elison @megelison.bsky.social · 21h
At this point, I assume any defense of AI was written by it. I’ve not been wrong.
218743
Reposted by Daniela Tafani
Antonio E. Porreca 🐳 @aeporreca.org · 27/09/2026
Imagine replacing people with tokens for plagiarism slot machines. This is why I take the extreme position that LLM-fuelled papers are not research, should not be read or cited, and should definitely count in one’s CV, but negatively.
46420
Reposted by Daniela Tafani
Dr Rebecca Wragg Sykes @lemoustier.bsky.social · 27/09/2026
The problem isn't rogue #AI models, it's *negligent companies*. If a biomedical firm put their product in public water supply because they wanted to test it 'in the real world', they'd be treated as criminal at corporate & individual level. www.theguardian.com/technology/2...
theguardian.com
OpenAI halts training of latest models as reports mount of AI agents going rogue
Decision follows disclosures that OpenAI agents searching government websites had acted in unexpected ways
20814321
Reposted by Daniela Tafani
Dwayne Monroe @cloudquistador.eurosky.social · 19/09/2026
Precisely. My message to my fellow US'ians, among the most heavily propagandized people on Earth, is to see the tech industry in its totality: military, capital, i.e. the political economy. I suggest starting here: yashalevine.com/surveillance...
yashalevine.com
Surveillance Valley
Surveillance Valley, a book about the secret military history of the Internet. In Surveillance Valley, Yasha Levine traces the history of the internet back to its beginnings as a Vietnam-era tool for ...
0233
Reposted by Daniela Tafani
Dwayne Monroe @cloudquistador.eurosky.social · 19/09/2026
It is surely no accident that, at this moment, when the global system established after WW2, (imperialist, with a liberal patina of 'learning lessons') is collapsing, we are encouraged to 'debate' about mythical thinking machines.
072
Reposted by Daniela Tafani
Dwayne Monroe @cloudquistador.eurosky.social · 26/09/2026
As her essay unfolds, Weil takes the reader on a tour of The Iliad, showing how what seems ancient, is in fact ever present. We never left the age of force, only the weapons have changed. /end
142
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 26/09/2026
Oh look, it is brainlike! Must be intelligent 🤔
walnut
722246
Reposted by Daniela Tafani
Shannon Vallor @shannonvallor.bsky.social · 25/09/2026
‘We’re going to lose some fine intellectual dexterity.’ This is a CEO openly confessing to a level of acquired mental indolence that, were he not coddled 24/7 by an army of paid underlings, would make him incapable of living independently and feeding himself. He’s fine with it. Your kids won’t be.
7340105
Reposted by Daniela Tafani
Fabio Chiusi @fabiochiusi.bsky.social · 25/09/2026
“The AI doomsday rhetoric not only helps the tech industry, it allows media outlets to chase clicks without actually challenging wealth and power.” 🎯
041
Reposted by Daniela Tafani
Sahil Kapur @sahilkapur.bsky.social · 25/09/2026
👀 OpenAI’s A.I. Tried to Breach 4 Other Targets, Without Prompting www.nytimes.com/2026/09/23/t...
85433
Reposted by Daniela Tafani
By Cory Doctorow (GPG 0xBF3D9110957E5F4C) @doctorow.pluralistic.net · 24/09/2026
...and at the end of it John Henry is like 'well, it felt kind of efficient using two hammers but I’m not sure what the point was.'" -Rusty Foster, Today in Tabs www.todayintabs.com/p/meta-s-new...
todayintabs.com
Meta's New A.I. Agent For Losers
Also: "fast as hell and built for sickos?" That's what they called me in high school.
49113
Reposted by Daniela Tafani
Tom Roberts @tpgroberts.bsky.social · 24/09/2026
I do think you basically charge a few of them with the cybercrimes they're committing and the "alignment" problem does suddenly start to disappear.
261454341
Reposted by Daniela Tafani
Dr. Damien P. Williams, dread portent down from a mountain cave @wolven.blacksky.app · 23/09/2026
Every time one of these companies rolls out a new "amazing" "AI" feature, it turns out to just be howevermany exploited humans in a trenchcoat. Every fucking time.
718968
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 23/09/2026
New depiction just dropped 🙃 mu.social/profile/ethi...
1404
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 19/09/2026
All we can expect when trying to scale up AI systems is diminishing returns and devastating environmental damage. Calls for `slowing down’ can conveniently hide the diminishing returns, while still convincing people that the tech is “all powerful”.
514756
Reposted by Daniela Tafani
Tyler @electriceden92.bsky.social · 22/09/2026
Fantastic thread. @olivia.science and others do a wonderful job of supporting your arguments here, including dismantling the mythos implied by the 'black box' zenodo.org/records/2007...
zenodo.org
Understanding Artificial Neural Networks: Mysterianism about Known Mechanism is Mysticism
Mysterianism is the idea that human cognition, mind, cannot be understood. Taking this concept and applying it to known mechanism — such that claims are made that we do not know how engineered systems...
2145
Reposted by Daniela Tafani
Antonio E. Porreca 🐳 @aeporreca.org · 19/09/2026
I don’t work on stuff suitable for STOC, but let me say that the 2027 edition LLM policy of one of the (former, AFAIC) “top” TCS conferences is pure shit, and I certainly don’t regret not being part of a community actively working on destroying science from the inside. Shame, shame, shame on you.
• Authors may use large language models (LLMs) and other generative AI tools in preparing papers.
Substantive use must be disclosed in the paper; minor copy-editing and grammar or clarity improvements to the authors' own text do not require disclosure, as detailed below.
• Program committee (PC) members and external reviewers (sub-reviewers) may use LLMs to assist with reviewing, with author consent required as described below. All reviews and decisions remain the responsibility of the PC members and sub-reviewers.
• STOC 2027 is also exploring private LLM-based tools to assist authors before submission and reviewers during evaluation; details will be announced later.Use of LLMs in reviewing and evaluating papers: PC members and sub-reviewers may use LLMs to assist in reviewing and evaluating the public ar Xiv version of a paper. Until that version becomes publicly available, they may use LLMs on the submitted PDF. Authors must explicitly consent to both uses as part of the submission process. This permission concerns the paper itself; reviewer reports and program committee deliberations remain confidential. Reviewers remain responsible for their reviews, and all decisions about papers will be made by humans. PC members and sub-reviewers must disclose all AI usage in preparing the review.

STOC 2027 is exploring private LLM-based tools to help authors review papers before submission and to assist PC members and sub-reviewers in evaluating submissions.
Details will be released closer to the submission deadline.
44213
Reposted by Daniela Tafani
🌈 Hailey Piper 🌈 @haileypiperfights.bsky.social · 17/09/2026
NBC News interviewed a plank with a face drawn on it and asked if it could fall in love with humans
Plank from Ed Edd n Eddy
823654
Daniela Tafani @danielatafani.bsky.social · 15/09/2026
"Un résultat sans chemin" Comme le dit @mcpbsk.bsky.social, appeler un taxi pour se rendre à l'arrivée d'un marathon, c'est ne pas avoir compris ce qu'est un marathon. @aeporreca.org
010
Reposted by Daniela Tafani
Joanna Bryson @j2bryson.bsky.social · 13/09/2026
"AI companies will inevitably face a large number of lawsuits as their products become more ingrained… When the legal wrangling begins, the industry’s extravagant (and strangely lighthearted) talk about the chance their agents could bring about human extinction won’t play well in court." #aiethics
wapo.st
Opinion | Why the Sam Bankman-Fried case should spook the AI elite
The fallen founder’s doomsday ideology helped put him behind bars.
0143
Reposted by Daniela Tafani
Antonio E. Porreca 🐳 @aeporreca.org · 14/09/2026
Not liking either the Leiden Declaration (weak), nor the high nobility statement (extremely weak for so-called “leaders” as well as hypocritical, Tao having been a strong and naive proponent of AI til 10 minutes ago), I suggest ahmath.org, which actually includes the crazy option of not using AI.
12610
Reposted by Daniela Tafani
Thom Scott-Phillips @thomscottphillips.bsky.social · 13/09/2026
To re-phrase the declaration: generating experimental findings in psychology is only a tool and proxy for achieving the primary goal of conceptual understanding and insight. Forgetting this in a world of advanced methodological tools may turn the tools against the primary goal.
1138
Daniela Tafani @danielatafani.bsky.social · 13/09/2026
"Feynman, commenting on how managerial and marketing decisions overrode engineering recommendations, leading to an avoidable calamity, stated that “For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.""
020
Reposted by Daniela Tafani
Olivia Guest · Ολίβια Γκεστ @olivia.science · 10/09/2026
It's pretty much illegal everywhere these companies operate. Yup. UK example: www.legislation.gov.uk/ukpga/1990/1...
legislation.gov.uk
Computer Misuse Act 1990
An Act to make provision for securing computer material against unauthorised access or modification; and for connected purposes.
315540
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 13/09/2026
"Using some of my own work, I set out to clear up a few relevant concepts from theoretical computer science and computational cognitive science that can help to build conceptual hygiene and immunity against hyped claims about “AI” capabilities." irisvanrooijcogsci.com/2026/09/12/c...
irisvanrooijcogsci.com
Computational complexity for hype immunity
Using some of my own work, I set out to clear up a few relevant concepts from theoretical computer science and computational cognitive science that can help to build conceptual hygiene and immunity…
917467
Reposted by Daniela Tafani
Teresa Heffernan @tjheffernan.bsky.social · 13/09/2026
Banting & Best, grateful for the resources of a public institution, sold the patent for insulin to the University of Toronto for $1 each. Banting: “Insulin does not belong to me, it belongs to the world.” He wanted all who needed it to have it. Big Pharma, Big Tech etc have destroyed that model.
085
Reposted by Daniela Tafani
Katie Hinde @mammalssuck.bsky.social · 12/09/2026
"The whole “Oh no, our models went rogue” thing … When a bridge collapses, you don’t analyze whether the bridge was ethical or sentient or why it decided to collapse. You ask, who is the person who built this bridge to be so flimsy?" www.wired.com/story/one-of...
wired.com
One of AI’s Fiercest Critics Says All the Doom Talk Is ‘Meant to Distract Us’
Timnit Gebru argues that AI companies are stoking fears of extinction to avoid discussing actual harms, like autonomous weapons.
302666905
Reposted by Daniela Tafani
Rania Combs @raniacombs.bsky.social · 12/09/2026
Exactly!
“If you showed up at Defcon and gave talk about how your autonomous malware did something unexpected and damaged someone else's computers, the first question from the audience would be "Why are you so shit at making secure sandboxes?" It wouldn't be "How are you so awesome at making hacking tools?"
09527
Reposted by Daniela Tafani
Barbara Fantechi @barbarafantechi.bsky.social · 11/09/2026
everyone considering this, have a look at ahmath.org compare and contrast
ahmath.org
AHM
0549
Daniela Tafani @danielatafani.bsky.social · 11/09/2026
@mariuzzoandrea.bsky.social
020
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 11/09/2026
Remember: doing math is a *cognitive capacity* mu.social/profile/iris...
0296
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 11/09/2026
No the plagiarism machine cannot do math, it can only steal math. Let mathematicians just do their work in peace, and we can gain deep new knowledge ✨
47620
Reposted by Daniela Tafani
Emily M. Bender @emilymbender.bsky.social · 11/09/2026
They can't, because the Doomer narrative *is* the Booster narrative. To debunk the Doomer stuff (and it *is* bunk), you have to point out that it's based on pure fantasy that the mathy maths are going to combust into consciousness ... >>
224335
Reposted by Daniela Tafani
xX💖🌹Xx_dolly_Angelic_xX🌹💖Xx @dollyangelic.bsky.social · 10/09/2026
So, I'm an academic and have worked with sensitive qualitative data that is to remain private and anonymous. I also work in an open plan office. Always-on recording in the office would violate my ethics approval. So my workplace should ban this shit from campus.
26027
Reposted by Daniela Tafani
Trevor A. Branch @trevorabranch.bsky.social · 08/09/2026
Millennium math problem solved by OpenAI may have "regurgitated" competitors unpublished work... also they spent more than $22 million in computing power in one week to scoop them. techcrunch.com/2026/09/08/o...
techcrunch.com
OpenAI fought dirty on career-making math problem, says NYU mathematician | TechCrunch
There is a $1 million bounty for the first person providing a solution to the Navier-Stokes existence and smoothness problem.
391331459
Reposted by Daniela Tafani
Melanie Mitchell @melaniemitchell.bsky.social · 08/09/2026
Hey @ericlipton.nytimes.com 👋 AI is not “evolving” on its own. *People* are choosing how to build it. *People* are choosing how to train it. *People* are choosing how to run it. Maybe the problem is that they are the *wrong people* to make these choices.
28635174
Reposted by Daniela Tafani
David Sepkoski @hallucigenia.bsky.social · 06/09/2026
Will someone please teach science journalists at the NYT that human behavioral genetics is 99% bullshit? A better headline for this piece would be "Psychologists Posing as Geneticists Played With Some Statistical Software." www.nytimes.com/2026/09/03/s...
nytimes.com
Scientists Find More Than 1,000 Genetic Variants Linked to Personality
More than one million genomes helped identify DNA variants linked to extroversion, agreeableness, neuroticism and more. Researchers measured how those traits influenced how people lived.
1014944
Reposted by Daniela Tafani
The Washington Post @washingtonpost.com · 04/09/2026
The two largest public school systems in the country, New York and Los Angeles, are banning AI for at least some of their students this school year.
washingtonpost.com
Another big school district bans generative AI, this time for all students
The action in the Los Angeles Unified School District comes after officials adopted strict screen time limits.
111024
Reposted by Daniela Tafani
Olivia Guest · Ολίβια Γκεστ @olivia.science · 24/02/2026
when people talk about AI used in moderation of social media posts, you must remember there is always the human-in-the-loop, often a woman, often in the Global South, always dehumanised and forgotten 1/n www.theguardian.com/global-devel...
theguardian.com
‘In the end, you feel blank’: India’s female workers watching hours of abusive content to train AI
Women in rural communities describe trauma of moderating violent and pornographic content for global tech companies
41049623
Reposted by Daniela Tafani
Olivia Guest · Ολίβια Γκεστ @olivia.science · 17/11/2025
Non-industry compromised scientists keep saying these models don't become safe, no matter what, but people keep thinking just because the concept of guardrails is mentioned it must work. By definition, it doesn't. This is not something open to discussion, unless you're a paid shill.
9675282
Reposted by Daniela Tafani
Olivia Guest · Ολίβια Γκεστ @olivia.science · 02/09/2026
When people adopt the logic of AI — like using superficial tells to accuse others of using AI, for example — they are demonstrating they are happy to use correlationism, conspiratorial thinking, and generally AI-like logic to think. So even if ostensibly they claim to be anti AI, they fall for it!
37720
Reposted by Daniela Tafani
Mother Jones @motherjones.com · 02/09/2026
New York City Mayor Zohran Mamdani announced a one-year moratorium on AI use for elementary and middle school students on Wednesday. Kids “need to wrestle with tough problems on their own,” New York’s mayor said Wednesday.
motherjones.com
Mamdani issues AI moratorium in NYC schools
Kids "need to wrestle with tough problems on their own,” New York's mayor said Wednesday.
15674183
Reposted by Daniela Tafani
aaron @aaron.mn · 02/09/2026
Weird how every AI generated video was created to put people out of work, spread misinformation, or do sex crime. Probably a coincidence. I'm sure we'll find a use for it eventually.
016747
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 02/09/2026
💯
092
Reposted by Daniela Tafani
Antonio E. Porreca 🐳 @aeporreca.org · 15/08/2026
You can register for attending Max Weinreich’s talk “The crisis of AI-generated mathematics” via Zoom on 11 Sep 2026, 1:20 ET at www.math.fsu.edu/~bauer/AI-Ma.... (Abstract²: I present the case for total opposition to the use of artificial intelligence in mathematics doi.org/10.48550/arX...)
math.fsu.edu
Register — The crisis of AI-generated mathematics
Register for Max Weinreich's talk, The crisis of AI-generated mathematics, in the Math and AI series.
083
Reposted by Daniela Tafani
Dr Abeba Birhane @abeba.blacksky.app · 02/09/2026
👏🏾👏🏾 techcrunch.com/2026/08/31/m...
techcrunch.com
MapQuest's app surges to No. 1 in Navigation after refusing to rename Lake Ontario | TechCrunch
MapQuest is getting an unexpected boost from its decision to defy President Trump’s renaming of Lake Ontario.
38014
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 01/09/2026
In case of use for others, sharing this slide in my intro class for a Bachelor course that I teach to AI students (and other interested students). More resources in thread 🧵
No (generative*) AI use allowed

Several (not mutually exclusive) reasons

AI use violates standards of scientific integrity.

AI use violates the examination and education rules (EER). 

This course is a skills course. Your task is to learn to read critically, think critically, model critically all by yourself. 

For more info see course manual. 

---
*But see: Guest, O., Suarez, M., et al. (2026). Against the Uncritical Adoption of 'AI' Technologies in Academia. Digital Culture & Education.
319653
Daniela Tafani @danielatafani.bsky.social · 31/08/2026
From a political point of view, this is a denial of health care - and the diversion of its financial resources to military companies - disguised as technological innovation
010
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 29/08/2026
bsky.app/profile/tjhe...
0122
Reposted by Daniela Tafani
Iris van Rooij 💭 @irisvanrooij.bsky.social · 29/08/2026
The conclusion is NOT "we need to debias LLMs" Besides that it's impossible, it also detracts from the point, which is: We shouldn't use machines to offload of the very cognition that is constitutive of the work to be done & supercharge existing inequities and injustices in the process. 🧵
544191