Sign in

josef-p.bsky.social

@josef-p.bsky.social
41 followers 173 following 0 posts
PostsRepliesMedia
Reposted by @josef-p.bsky.social
Maddy Thorson @maddymakesgames.com · 05/10/2026
one thing you learn from making a lot of things is that the nitty gritty details, the millions of tiny invisible decisions, ARE the art. The "higher level ideas" are not nearly as important or interesting
252074500
Reposted by @josef-p.bsky.social
Emily M. Bender @emilymbender.bsky.social · 19/09/2026
But speaking of staying in one's lane, can we PLEASE switch the default position from "It's fine to use 'AI' for this thing I'm not an expert in" to "Yeah, I'd be really skeptical of any use cases and need to see strong evidence that it's okay, but I'm not the one who can produce that evidence"?
12569154
Reposted by @josef-p.bsky.social
Dr Abeba Birhane @abeba.blacksky.app · 02/08/2026
"Our role as academics is to protect this one institution. This is why we receive tax money; why we're basically civil servants. The role of the university is even lost on many, so bears repeating: produce knowledge and educate at the highest level. AI destroys both at their core"
519161
Reposted by @josef-p.bsky.social
Robert Evans (the Only Robert Evans) @iwriteok.bsky.social · 25/07/2026
bloomberg has a fun article today arguing that the Cybertruck is the biggest flop in automotive history. it includes this glorious graph.
A graph showing how the edsel and Cybertruck both performed based on their sales target and actual year one and year two sales. The edsel was expected to sell 200k units in its first year. It sold slightly over 50k, and less than 50k the next year

The Cybertruck had a sales.target of 250k in its first year. They sold around 40k, the next year they sold about half as many.
416159783945
Reposted by @josef-p.bsky.social
Jan Lehnardt @janl.bsky.narrativ.es · 22/07/2026
I’m surely not the first to come to this realisation, but the LLM industrial complex is to tech-life what MAGA is to politics: erode any and all trust in established institutions, people and practices for the benefit of the very few and worst people on the planet in the (poor) guise of freedom.
312443
Reposted by @josef-p.bsky.social
Naomi Klein @naomiaklein.bsky.social · 02/07/2026
Everybody in the "climate space" (which means everyone on earth...) needs to understand these mind-blowing figures. Google's "total electricity consumption jumped from 31 terawatt hours (TWh) in 2024 to 43 TWh in 2025." In ONE year. The AI corporate arms race is eating the world.
7043292446
Reposted by @josef-p.bsky.social
Paris Marx @parismarx.com · 09/06/2026
”Meta, Spotify, and Google don’t just host AI-generated imagery, ads, and music; they’re also responsible for making the tools that create it. … Allowing users to filter it out regardless would go against all the effort … to profit from AI: They want you to embrace the slop factory.”
theverge.com
Let us filter AI slop, you cowards
Let’s put those AI labels to the test.
113224
Reposted by @josef-p.bsky.social
Ed Zitron @edzitron.com · 27/06/2026
I love watching grown men walk around saying stuff like "p doom" and "permanent underclass." True cargo cult shit, an insane way to talk that sounds weird and exists only to signal to other peculiar people that you know the special words that bring about the great capitalist prophecy
1674367
Reposted by @josef-p.bsky.social
tante @tante.cc · 26/06/2026
It's weird how every little automation is an "agent" these days. Weird LLM that tags your uploaded photo: Agent. A sequence of filters and pipes in a shell: Agent. Pre-saved prompt: Agent. "AI" burns through terms quicker than even crypto did.
107811
Reposted by @josef-p.bsky.social
Emily M. Bender @emilymbender.bsky.social · 23/06/2026
A phenomenon I've noticed recently is people trying to occupy some untenable middleground wrt to the use of systems sold as "AI" -- this is a position where people try to recognize the harms of this tech but also hold space for "responsible" or "ethical" use. 🧵>>
17598255
Reposted by @josef-p.bsky.social
Thomas Spooks 🎃👻💀 @thomasfuchs.at · 25/06/2026
To summarize, AI will cause personal computers to cost $10,000, all applications are frozen to about 2025 design and implementation, power to run the computer will be twice the price and also you need to sign in with your passport to start your computer. Explain to me again how this is progress?
1422744
Reposted by @josef-p.bsky.social
Emily M. Bender @emilymbender.bsky.social · 14/06/2026
I have been working on societal impacts of language technology since late 2016, and my nightmare example since pretty early on has been using automatic transcription in emergency services systems. And today I learn it's happening *in my own city* www.seattletimes.com/seattle-news... Short 🧵>>
seattletimes.com
Seattle uses AI to help triage, divert 911 medical calls
The Seattle Fire Department started using AI technology during live 911 calls in 2023. There hasn't been any public review process.
22917434
Reposted by @josef-p.bsky.social
Christian Ward @cjward.bsky.social · 10/06/2026
"We already live in a world where you walk outside and there are billboards and signs that are obvious AI slop. That’s become part of our visual reality. To me, generative AI feels less like innovation than a symptom of a broader cultural and economic rot." I like this kid.
1876992123
Reposted by @josef-p.bsky.social
Yvonne Lam @yvonnezlam.bsky.social · 25/05/2026
Tangentially, it continues to bug me that people are willing to structure information for machines that they weren't willing to structure for other people, e.g. coworkers.
25915
Reposted by @josef-p.bsky.social
Ed Zitron @edzitron.com · 24/05/2026
Wrote up this story - honestly it undersells how bad it is. OpenAI had a negative 122% non-GAAP operating margin in Q1 2026 and ChatGPT growth has stalled. OpenAI is a company with broken economics that has hit its ceiling. www.wheresyoured.at/news-openai-...
News: OpenAl Had A Negative 122% Non-GAAP Operating Margin In Q12026, and ChatGPT
Growth Has Stalled
ED ZITRON / MAY 22, 2026 / 4 MIN READ
Executive Summary:
• The Information reports that OpenAl generated $5.7bn in revenue for the first quarter of 2026 based on discussions with sources familiar with its financials.
• With adjusted negative margins of -122%, this means that for every dollar of revenue OpenAl made, it lost an additional $1.22, or around $6.95bn on a non-GAAP basis.
• OpenAl is "on track" to hit goal of $30bn in 2026 revenue, but margins suggest losses of over $36.6bn.
• OpenAl continues to struggle converting free ChatGPT users to paying customers, and overall user growth has stalled.
942489
Reposted by @josef-p.bsky.social
Paris Marx @parismarx.com · 20/05/2026
ChatGPT was never Google’s biggest threat; it was always its own hubris. With Search now being sacrificed on the altar of AI dominance, the time has come for users to get serious about moving off Google services. It’s not only destroying its products; it’s trying to further degrade the web itself.
disconnect.blog
Google is its own worst enemy
In its bid to push AI, it’s destroying its products and trying to kill the web too
221562524
Reposted by @josef-p.bsky.social
Iris van Rooij 💭 @irisvanrooij.bsky.social · 10/05/2026
Meet the academics refusing to use generative AI Researchers say they have their reasons for avoiding AI tools — and they’re sick of arguing about it. www.nature.com/articles/d41... h/t @maartenp.bsky.social
nature.com
Meet the academics refusing to use generative AI
Researchers say they have their reasons for avoiding AI tools — and they’re sick of arguing about it.
26905354
Reposted by @josef-p.bsky.social
G. Willow Wilson @gwillow.me · 04/05/2026
This is for *your* kids, you understand. The tech elites send *their* kids to private, screen-free Montessori schools. They know exactly what they're doing.
4944221846
Reposted by @josef-p.bsky.social
Cominsitu @cominsitu.bsky.social · 08/04/2026
the thing you don’t realize about the apocalypse is that you basically have to keep working through it
510523
Reposted by @josef-p.bsky.social
Dr. Jens Foell @jensfoell.de · 15/04/2026
OpenAI: We’re burning money like the Joker. A miracle needs to happen for us to turn a profit Microsoft: Please please use our AI systems, we’re teetering on the edge here Anthropic: I wonder what’ll kill us first: lawsuits, regulations or model collapse Media and universities: AI is here to stay
120111864070
Reposted by @josef-p.bsky.social
Rua M. Williams @fractalecho.bsky.social · 12/04/2026
Very unimpressed by pop neurodiversity influencers and their AI apologia. More bullshit claiming that anti-AI stances are classist and ableist and racist. Meanwhile, AI is built on exploited black labor, kills disabled people every day, disposesses the working class, and drinks all our water.
7387113
Reposted by @josef-p.bsky.social
Brian Merchant @bcmerchant.bsky.social · 26/03/2026
This is the way
404media.co
Wikipedia Bans AI-Generated Content
“In recent months, more and more administrative reports centered on LLM-related issues, and editors were being overwhelmed.”
141011202
Reposted by @josef-p.bsky.social
404 Media @404media.co · 26/03/2026
NEW: Wikipedia has banned AI-generated content.
404media.co
Wikipedia Bans AI-Generated Content
“In recent months, more and more administrative reports centered on LLM-related issues, and editors were being overwhelmed.”
201237816753
Reposted by @josef-p.bsky.social
Emily M. Bender @emilymbender.bsky.social · 21/03/2026
There's a learned helplessness here that feels a lot like obeying in advance. Use your brain, don't let anyone tell you you can't, and definitely don't tell others they can't.
015826
Reposted by @josef-p.bsky.social
Amyville Hoyrror @amyhoy.bsky.social · 22/03/2026
machine translation is acceptable for "help me sort of make sense of this news article" or "plz help the instruction manual isn't in my language" or "only an obscure blog in japan ever covered this vintage lens…" not for anything that is meant to endure
1506
Reposted by @josef-p.bsky.social
Ed Zitron @edzitron.com · 17/03/2026
One other note from today's newsletter: I'm worried about over-use of LLMs in software engineering. If people aren't reading (or are barely reading) code they ship, "writing code fast" becomes a liability. It also can stop you from learning to solve problems. www.wheresyoured.at/why-are-we-s...
I also want to be clear that I am extremely nervous about how many people appear to be fine with not reading code. I am currently (very slowly) learning Python, and every new thing I learn reinforces my overwhelming anxiety that there is a lot of software being written today by people who don’t read the output from LLMs and, in some cases, may not have understood it if they did. While I’m not saying all or even many software engineers might do this, I am alarmed by the idea that it’s becoming more commonplace — and even more alarmed that the reaction appears to be “ah it’s fine who gives a shit, it works.” 

Guess what! It doesn’t always work. Amazon Web Services had multiple recent outages caused by use of its Kiro AI coding tool, and while it insists that AI isn’t to blame, it also convened an internal meeting to discuss this specific issue, and The Financial Times reported that Amazon now requires junior and mid-level engineers to get sign off on AI-assisted changes to code. However you may feel about Amazon as a service, its engineers are likely indicative of corporate engineering on some level, which is making me wonder if we’re not going to have some real problems in software development in the next few years as a result.

What does the software industry look like if nobody is actually reading their code? How many software engineers are comfortable doing this? I’m sure somebody will read this and get terribly offended, but to be clear, I’m not accusing you of copy-pasting code you can’t understand and being happy if it works unless that’s exactly what you’re doing. 

To be explicit, allowing LLM to write all of your code means that you are no longer developing code, nor are you learning how to develop code, nor are you going to become a better software engineer as a result. 

This isn’t even an insult or hyperbole. If you are just a person looking at code, you are only as good as the code the model makes, and as Mo Bitar recently discussed, these models are built to galvanize you, glaze you, and tell you that you’re remarkable as you barely glance at globs of overwritten code that, even if it functions, eventually grows to a whole built with no intention or purpose other than what the model generated from your prompt. 

I’m sure there are software engineers using these models ethically, who read all the code, who have complete industry over it and use it like a glorified autocomplete. I’m also sure that there are some that are just asking it to do stuff, glancing at the code and shipping it. It’s impossible to measure how many of each camp there are, but hearing Spotify’s CEO say that its top developers are basically not writing code anymore makes me deeply worried, because this shit isn’t replacing software engineering at all — it’s mindlessly removing friction and putting the burden of “good” or “right” on a user that it’s intentionally gassing up.

Ultimately, this entire era is a test of a person’s ability to understand and appreciate friction. 

Friction can be a very good thing. When I don’t understand something, I make an effort to do so, and the moment it clicks is magical. In the last three years I’ve had to teach myself a great deal about finance, accountancy, and the greater technology industry, and there have been so many moments where I’ve walked away from the page frustrated, stewed in self-doubt that I’d never understand something.

I also have the luxury of time, and sadly, many software engineers face increasingly-deranged deadlines set by bosses that don’t understand a single fucking thing, let alone what LLMs are capable of or what responsible software engineering is. The push from above to use these models because they can “write code faster than a human” is a disastrous conflation of “fast” and “good,” all because of…
1735367
Reposted by @josef-p.bsky.social
Ed Zitron @edzitron.com · 14/02/2026
Hello everyone! I am working on something and could use your help. If you pay for Claude code, please type npx ccusage@latest into your terminal and send me a screenshot of your usage from the last month, along with your level of subscription ($20/100/200). I would be so grateful!
2637697
Reposted by @josef-p.bsky.social
Jason Koebler @jasonkoebler.bsky.social · 10/02/2026
Ring is a wildly dangerous company. Always has been. But it has sort of flown under the radar the last couple years as it tried to soften its image. Make no mistake that this is an extremely dangerous surveillance dragnet: www.404media.co/with-ring-am...
404media.co
With Ring, American Consumers Built a Surveillance Dragnet
Ring's 'Search Party' is dystopian surveillance accelerationism.
5427221037
Reposted by @josef-p.bsky.social
Emily M. Bender @emilymbender.bsky.social · 06/02/2026
But now we are fortunate to have @fractalecho.bsky.social 's amazing book out in the world. Read it to be better prepared to recognize & combat ableism and eugenics in "AI" discourse. link.springer.com/book/10.1007...
link.springer.com
Disabling Intelligences
This book discusses the influences of eugenics on the AI industry and the impacts of AI opportunism on disabled people.
19922
Reposted by @josef-p.bsky.social
Luke Plunkett @lukeplunkett.com · 15/01/2026
im so tired
331450
Reposted by @josef-p.bsky.social
Novelty Cistern @noveltycistern.bsky.social · 21/12/2025
This and the discussion about how academics can now just read AI summaries of things really drives home how much these guys always confuse the product for the goal. You don't do a coloring sheet because you want to have a colored sheet at the end!!! The coloring is the point!!!
353000639
Reposted by @josef-p.bsky.social
Benjamin Kunkel @kunktation.bsky.social · 17/04/2025
I wrote about an obscure German contribution to the critique of political economy, for @harpers.bsky.social harpers.org/archive/2025...
harpers.org
The Sameness of Different Things, by Benjamin Kunkel
Reading a new translation of Capital
24013
Reposted by @josef-p.bsky.social
Jathan Sadowski @jathansadowski.com · 04/10/2025
The AI bubble is gonna tank the economy. The ironic future of AI is to throw us backwards into a great depression. Let’s gooooo www.marketwatch.com/story/the-ai...
114434
Reposted by @josef-p.bsky.social
Hypervisible @hypervisible.blacksky.app · 05/10/2025
I think often about the Amazon engineer who wrote in an open letter to the company that Ring was “incompatible” with a free society.
washingtonpost.com
Analysis | Amazon’s Ring plans to scan everyone’s face at the door
For the first time, the company is putting facial recognition into its home security doorbells and video cameras.
624093
Reposted by @josef-p.bsky.social
C. L. Polk @clpolk.com · 28/09/2025
Coates says Black Americans have stories about relatives who were lynched or had to flee and that a life free of political violence is not in reach in his time Klein: "Sometimes I think that having a historical scope that wide can make the present too deterministic." They lynched a kid last week.
411683505
Reposted by @josef-p.bsky.social
dash @knowpicnic.bsky.social · 23/07/2025
This paragraph by @edzitron.com is everything. www.wheresyoured.at/the-haters-g...
5479133