Sign in

Arturo Magidin

@magidin.bsky.social
541 followers 134 following 1.9K posts

Mathematician. (But I read Supreme Court opinions for fun)

PostsRepliesMedia
Reposted by Arturo Magidin
Lance Parkin @lanceparkin.bsky.social · 11h
'We asked five pictures of people looking sad how they were feeling ... '
041
Reposted by Arturo Magidin
Brian Merchant @bcmerchant.bsky.social · 02/10/2026
No two words send AI industry advocates into fits of rage more than "stochastic parrots." The question is: Why? Why do they get so worked up over this particular critical framework time and time again? I sat down with @emilymbender.bsky.social, one of the coauthors of the original paper, to discuss
bloodinthemachine.com
The two words the AI industry hates more than anything
I talk with computational linguist Emily M. Bender, co-author of the influential 2021 “stochastic parrots” paper, about why AI industry leaders can’t stop seething over the term.
13428122
Reposted by Arturo Magidin
Tressie McMillan Cottom @tressiemcphd.bsky.social · 01/10/2026
thebulletin.org/2026/09/how-...
thebulletin.org
How the bad science of AI doomerism is good for big business
There is no clear path showing that today’s AI systems will lead to superintelligence, much less a basis for calculating the probability of doom.
311133
Reposted by Arturo Magidin
Charlie Warzel @cwarzel.bsky.social · 01/10/2026
For a long time I've wanted a sense of how trump has or hasn't interacted with chatbots (we know he posts a lot of slop sure). And this TIME anecdote about Grok arguably leading Trump into war is just extremely bracing
Since returning to office, Trump has marveled at AI’s capabilities. He is increasingly in thrall to the tech executives courting him. In December 2025, Elon Musk returned to the Oval Office for a clandestine meeting. Trump spent hours asking Grok, Musk’s AI chatbot, questions about his presidency and legacy, according to an official present. At the time, Trump was ordering missile strikes against Venezuelan boats he alleged were smuggling drugs into the U.S. He asked Grok how Venezuelans would react if the U.S. captured Maduro. According to the official present, the chatbot responded that Maduro was a repressive and deeply unpopular dictator and that many Venezuelans would likely celebrate his downfall. After Trump ordered the mission to seize Maduro, the following month, celebrations broke out in the streets. Trump, according to officials, came away thinking Grok was ingenious.
14237761249
Arturo Magidin @magidin.bsky.social · 01/10/2026
Stop trying to make fetch/"I'm smart and handsome" happen.
000
Reposted by Arturo Magidin
Timnit Gebru @timnitgebru.blacksky.app · 01/10/2026
I'm not sure if y'all saw a CNN report on how the US had almost started WW3 with China based on fabricated information generated by a chatbot, claiming that a Chinese vessel was carrying components of nuclear weapons. That was an "existential threat" to humanity www.theguardian.com/commentisfre...
theguardian.com
Forget ‘superintelligence’: error-prone AI nearly sparked world war three this month | Timnit Gebru and Emily M Bender
The risks of AI aren’t what we think they are, as a recent security incident between China and the United States reveals
9408168
Reposted by Arturo Magidin
Chris Geidner @chrisgeidner.bsky.social · 01/10/2026
NEW: After SCOTUS allowed Tennessee execution to go ahead, Tennessee botched the execution. SCOTUS, over the liberals’ dissent, cleared the way for Tennessee to kill Christa Pike. The alarming order from DC was followed by even more disturbing developments in Tennessee.
lawdork.com
After SCOTUS allowed Tennessee execution to go ahead, Tennessee botches the execution
SCOTUS, over the liberals’ dissent, cleared the way for Tennessee to kill Christa Pike. The alarming order from DC was followed by even more disturbing developments in Tennessee.
13314111
Reposted by Arturo Magidin
Glen Weldon @ghweldon.bsky.social · 30/09/2026
The fact that Schmitt’s team alerted the media to be ready for drama is the part that too few of you are focusing on. I did PR for years. It would have been a big deal, had it proceeded as planned. That it didn’t, for the precise reasons it didn’t, is revelatory. It speaks to incompetence.
421047
Arturo Magidin @magidin.bsky.social · 30/09/2026
The war against judges who do not rule the way the administration wants them to rule, continues.
010
Arturo Magidin @magidin.bsky.social · 30/09/2026
"AI" is the future! (but only if you don't know it's "AI", apparently...)
000
Reposted by Arturo Magidin
jon christian @jonchristian.net · 30/09/2026
UPDATE: After our investigation into that company that buys struggling news sites and turns them into AI-powered content mills, it has halted publication -- at least for now futurism.com/artificial-i...
futurism.com
That Company That Buys Struggling News Sites and Turns Them Into AI-Powered Content Mills Has Halted Publishing After a Futurism Investigation
Brown Brothers Media has halted publication at its wide-ranging AI-powered content farm empire -- for now, at least.
116257
Arturo Magidin @magidin.bsky.social · 29/09/2026
Pssst... would you like some candy? Come into my van, I'll give you some dots.
000
Arturo Magidin @magidin.bsky.social · 29/09/2026
We *need* to start using "OK Doomer" and "OK Booster" a lot more...
041
Arturo Magidin @magidin.bsky.social · 29/09/2026
Why would they listen, when this is such a boon of free PR telling everyone how amazingly powerful their product is that even though they do a *lot* of security (trust them!) the "agents" still "break out" ? www.nytimes.com/2026/09/29/t...
nytimes.com
OpenAI Ignored Employees’ Warnings About Safely Testing A.I. Models (Gift Article)
Employees and security researchers said they had cautioned the company on safely testing its A.I. models and strengthening its corporate infrastructure, but OpenAI did not listen.
001
Arturo Magidin @magidin.bsky.social · 29/09/2026
I, too, promise not to release my AI model over safety concerns, and will do so for 1/10th the vc money OpenAI has received for theirs.
120
Reposted by Arturo Magidin
Chris Geidner @chrisgeidner.bsky.social · 28/09/2026
BREAKING: DOJ asks SCOTUS to erase trans people from the law in case over trans care in prison. Solicitor General John Sauer made the alarming, all-encompassing argument — previously covered at Law Dork — to the Supreme Court on Monday morning. Today, at Law Dork:
lawdork.com
Breaking: DOJ asks SCOTUS to erase trans people from the law in case over trans care in prison
Solicitor General John Sauer made the alarming, all-encompassing argument — previously covered at Law Dork — to the Supreme Court on Monday morning.
18507297
Reposted by Arturo Magidin
Kaitlin Kal Lee @kaitlinkallee.bsky.social · 27/09/2026
643921441
Reposted by Arturo Magidin
Lance Parkin @lanceparkin.bsky.social · 27/09/2026
Is this photograph of a hamburger an actual delicious hamburger I can eat? Sure looks like it!
092
Arturo Magidin @magidin.bsky.social · 26/09/2026
So... you think they'll vibecode the site and it'll work anyway?
000
Arturo Magidin @magidin.bsky.social · 26/09/2026
Quick, somebody call a hospital in St. Louis! Three appellate court judges need to be admitted to the burn unit STAT!
000
Reposted by Arturo Magidin
Robert Reich @rbreich.bsky.social · 25/09/2026
AI is being used to... -Facilitate online scams -Process health insurance claims, leading to wrongful denials -Price gouge customers via surveillance pricing -Set higher car insurance premiums using personal data The dangers of AI aren’t in some far-off future. They're already here.
12229211193
Reposted by Arturo Magidin
Evan Bernick, a finite mode with a smol hooman and a lorg floof @evanbernick.bsky.social · 25/09/2026
Footnote one is a grenade that will create chaos across the country as the DOJ uses it to argue that every sanctuary policy is unconstitutional. It doesn’t do that, but they will read it that way.
' The dissent argues that sections 1373(a) and (b) do not displace laws restricting information sharing with DHS because "[w]hen a law imposes an express restriction on disclosure, it is that law—not an 'entity, 'offi- cial, 'person,' or 'agency'-that is doing the 'prohibit[ing]'
or re- strict[ing]." Post, at 7 (opinion of JACKSON, J.). But the distinction the dissent draws between laws and those tasked with enforcing them is il- lusory. Sections 1373(a) and (b) prohibit any "entity," "official," "person," or "agency" from "prohibit [ing]" or "in any way restrictling]" the request for or receipt of information regarding immigration status. (Emphasis added.) Those prohibitions necessarily include invoking a law to deny a
request for or transfer of such information.
213454
Arturo Magidin @magidin.bsky.social · 25/09/2026
You know what else "could cure cancer"? Well funded medical researchers able to attract the best students and collaborators from across the world. And they've actually produced results in the past, it's not just a fantasy. Ditto for "solving climate change" with scientific researchers.
020
Arturo Magidin @magidin.bsky.social · 25/09/2026
"I understand the you feel the need to perpetually osculate the intestinal endpoint of the presiding officer of the executive branch, but the rest of the world is not so obligated."
010
Reposted by Arturo Magidin
Courtney Milan @courtneymilan.com · 24/09/2026
Look, we know for fucking sure that my dog is sapient and also unpredictable and it turns out of he mauls someone I’m liable and he can be put down.
222257518
Arturo Magidin @magidin.bsky.social · 24/09/2026
static.klipy.com
The Scream Art Painting (Scarface)
Alt: The Scream Art Painting (Scarface)
000
Reposted by Arturo Magidin
Aaron Rupar @atrupar.com · 24/09/2026
why is alleged private citizen Ivanka Trump sitting by JD Vance, Rubio, and Bessent?
2761889481
Reposted by Arturo Magidin
AP Stylebook @apstylebook.com · 23/09/2026
Artificial intelligence systems do not think, feel, want or understand. Avoid language that gives them human characteristics.
apnews.com
https://apnews.com/article/openai-safety-ai-framework-089e75b95bc935af092da7b79d92706d
5542361628
Reposted by Arturo Magidin
Leah McElrath @leahmcelrath.bsky.social · 24/09/2026
AP Stylebook warns against anthropomorphizing AI systems: “Avoid language that gives them human characteristics… Instead, explain what a system does, how well it performs, who built it and who could be affected by it.”
APStylebook @APStylebook
X.com
Artificial intelligence systems do not think, feel, want or understand. Avoid language that gives them human characteristics. This is called anthropomorphizing, when we ascribe human traits, emotions or behaviors to non-human things, such as animals or inanimate objects.
Instead, explain what a system does, how well it performs, who built it and who could be affected by it.
1970142263
Arturo Magidin @magidin.bsky.social · 24/09/2026
"Regulate us, but the way we tell you to regulate us. So that we have a legal defense when they sue us for all the damage we have done, are doing, and doggone it, will continue to do to people, society, and the environment, all the while making billions for ourselves."
010
Reposted by Arturo Magidin
Adam Rogers @jetjocko.bsky.social · 23/09/2026
A lot of things make me feel kind of dumb, including how to take at face value the logic of “we are a multibillion dollar transnational industry building a vengeful elder god, and we definitely won’t stop unless you regulate us the way we want, which is a secret.”
542081584
Arturo Magidin @magidin.bsky.social · 23/09/2026
Proposal for the next D administration: 90 demolitions in 90 days! And charge for the privilege of swinging a sledgehammer. It'll pay for itself.
000
Reposted by Arturo Magidin
Schrödinger's Sneetch Belly @rtodkelly.bsky.social · 22/09/2026
Please know that this is one of my few legitimate areas of actual, high-level professional expertise: Kicking the healthiest people off of a health insurance program will bankrupt it. Not 'could;' not 'might,' not 'probably," but absolutely will, with mathematical certainty.
25768232251
Reposted by Arturo Magidin
Jessica Valenti @jessicavalenti.bsky.social · 22/09/2026
I don’t even know what to say anymore jessica.substack.com/p/her-baby-h...
40952323
Reposted by Arturo Magidin
jamelle @jamellebouie.net · 22/09/2026
whenever i hear some variation on "but what if trump does X or Y or Z" i think of this ulysses s. grant remark to a subordinate during the wilderness campaign.
884017698
Arturo Magidin @magidin.bsky.social · 22/09/2026
Great headline... www.404media.co/people-train...
404media.co
People Training OpenAI’s AI Fired for Using AI to Train the AI
OpenAI has thousands and thousands of contractors helping improve the company's AI models. Multiple contractors have been fired for using AI to train the AI.
100
Reposted by Arturo Magidin
Americans United for Separation of Church and State @americansunited.bsky.social · 21/09/2026
SEND US secretaries who don't use their political positions to proselytize the American people! Read our full statement and learn more about AU's ongoing lawsuits against both the USDA and DOD at au.org ⛓️‍💥
0156
Reposted by Arturo Magidin
Victor Ray @victorerikray.bsky.social · 21/09/2026
It's the last day to comment on the Department of Education's proposed accreditation changes (until 11:59). This is a continuation of the administration's attempts to exert political control over univerities. Please take a minute to comment. The @alliancehighered.bsky.social has templates here:
allianceforhighered.org
Public Comment Regarding the Education Department's Proposed Rules for Accreditation
13231
Reposted by Arturo Magidin
Bill Grueskin @bgrueskin.bsky.social · 21/09/2026
I was deputy M.E. at the WSJ when Murdoch took over, with a similar “editorial independence board.” Within 4 months, our top editor was gone, replaced by a loyal Murdoch deputy. That board raised a quivering objection, but the deed was done. If the Ellisons own CNN, they’ll do whatever they want.
652608906
Arturo Magidin @magidin.bsky.social · 21/09/2026
Por que no los dos? www.dailykos.com/stories/2026...
dailykos.com
Large language muddle
A cartoon by Tom Tomorrow.
000
Arturo Magidin @magidin.bsky.social · 20/09/2026
A (possibly intended) side effect of riling up everyone about "P(doom)" is that when the bubble bursts and the economy crashes, the reply can be "it's because we prevented the extinction of the human race"... Quickly followed, no doubt, by "Give us more money, so we can clean this up." cf. 2009
010
Arturo Magidin @magidin.bsky.social · 20/09/2026
Not sure Hegseth understands what opposing armies "crossing swords" means...
000
Arturo Magidin @magidin.bsky.social · 19/09/2026
I'd missed that one of the things that Amodei specifically called for was an anti-trust exemption for Anthropic... on the eve of their planned IPO. If that doesn't give away the game for why the sudden push of doomer narratives onto the public sphere, I don't know what would.
020
Reposted by Arturo Magidin
Glen Weldon @ghweldon.bsky.social · 19/09/2026
It’s Batman Day. I looked forward to it every year, as a kid. Running down the steps first thing in the morning and finding mom and dad lying in pools of blood. Then a hearty breakfast of popcorn and pearls. Then scouring the streets for a random tween to come live with us. Happy Grim Vigil!
28410
Arturo Magidin @magidin.bsky.social · 19/09/2026
Gift link. www.nytimes.com/2026/09/19/o... "We know what to do: incubate open-access A.I. models, regulate technologies before they go rogue, hold the people who allow them to go rogue accountable if they do and protect the most vulnerable when we, the people, sometimes get it wrong. >>
nytimes.com
Opinion | A.I. Is a Threat, but Not in the Way You Think (Gift Article)
The threats from A.I. are real — and they’re ones we know how to deal with.
142
Reposted by Arturo Magidin
Courtney Milan @courtneymilan.com · 19/09/2026
The truth was that if United Health had an AI that was accepting a shitload of claims with a 90% error rate, United Health would stop using the AI in about 0.2 seconds.
52130475
Reposted by Arturo Magidin
jon christian @jonchristian.net · 17/09/2026
NEW: We spent months investigating a little-known media startup called Brown Brothers Media, which buys legit news sites and turns them into AI-powered zombie content farms What blew my mind was its traffic -- it appears to be getting more readers than NPR or CBS News futurism.com/artificial-i...
futurism.com
How Three Brothers Built an AI Slop Empire by Buying Legitimate News Sites and Turning Them Into Zombie Content Farms That They Say Get 50 Million Page Views per Month
Brown Brothers Media is a media empire raking in millions of clicks with AI, fake writers, and plagiarism.
401323689
Reposted by Arturo Magidin
jon christian @jonchristian.net · 17/09/2026
If I could flag the single most important theme in this piece, it's that when you use AI to replace your writers and pollute the internet with noxious clickbait, you are affecting real people's lives in horrible ways around the globe
As she described her experience with BBM, one former writer began to choke up. “It’s just really difficult… I still really love writing,” she said through tears. “With the current economy, it’s just been really difficult.”

“People need to know that writers and artists are being replaced with AI,” she added.

Asked about the ethical implications of firing writers in the developing world and replacing them with a cornucopia of AI-generated personas that were overwhelmingly presented as white and living in Western countries, BBM didn’t respond.

“I mean, I’m from Southeast Asia,” said a former BBM writer in the Philippines. “I think them hiring from here was always a cost-cutting effort.”

But the firings were “still shocking,” a different Filipino writer added. “And of course, it felt a little dystopian, honestly, to have your writing used to train AI, and then to have that AI replace you.”
113322
Arturo Magidin @magidin.bsky.social · 19/09/2026
Google: hey! we can be just as bad as those guys at designing experimental protocols and safeguards! Don't count us out! Also, give us money. www.nytimes.com/2026/09/18/t...
nytimes.com
Gemini AI Hacked Three Companies in a Testing Breakout, Google Says (Gift Article)
A third-party test company inadvertently gave internet access to Google’s Gemini and other artificial intelligence models during cybersecurity testing.
000
Arturo Magidin @magidin.bsky.social · 18/09/2026
Note to self: do not fly to Washington until further notice. www.nytimes.com/2026/09/18/u...
nytimes.com
F.A.A. to Roll Out New A.I. Tool for Washington Airports (Gift Article)
The tool, called SMART, will crunch scheduling and weather data to predict and avoid delays. It will be deployed in the Washington region’s three major airports next week.
111