Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 1h@georgemonbiot.bsky.social looks like they killed metaphor along with irony 010 1
Reposted by Leo Ré Jorgetante @tante.tldr.nettime.org.ap.brid.gy · 13/10/2026Weird how the AI people are so scared of the "paperclip maximizer" destroying us while reshaping the whole economy around infrastructures for stochastic token machines. 118 8
Reposted by Leo Ré JorgeGlyph @glyph.im · 10/10/2026this is the perfect metaphor to understand the distinction between AI doomer concerns versus AI skeptic concernsfurry.engineer 0067 67
Reposted by Leo Ré Jorgetante @tante.tldr.nettime.org.ap.brid.gy · 03/10/2026So the "I left $AILab for ethical reasons" grift is starting up: 040 4
Reposted by Leo Ré JorgeGlyph @glyph.im · 29/09/2026Utilitarianism is broken because quality of experience is not actually a scalar value. It does not ideologically admit the possibility of the need for things like balance and rhythm. Consider the fact that you can die of thirst and you can die of drowning. While you *can* reflect this fact in a […]mastodon.socialOriginal post on mastodon.social 314 4
Reposted by Leo Ré JorgeGeorge Monbiot @georgemonbiot.bsky.social · 28/09/2026Here is how politics, in most cases, works: Rich and powerful interests, whether in fossil fuels, finance, big tech, media, meat, pharma or foreign governments, make a demand, and use their sophisticated tools to persuade a government to adopt it. The government adopts it. Please read whole🧵 31914398 914
Reposted by Leo Ré JorgeGeorge Monbiot @georgemonbiot.bsky.social · 18/09/2026A short thread on “legitimate concerns”. The phrase is used repeatedly by UK government ministers. The Home Secretary is at it again this week. But only ever - as far as I can tell - in one context: people’s response to the arrival of asylum seekers. Please read whole 🧵before responding - thanks 671280509 1.3K
Reposted by Leo Ré JorgeGlyph @glyph.im · 17/09/2026RE: framapiaf.org/@davidrevoy/117281447… unironically, unashamedly do this. call your friends, ask them for help. and when they say “wait did you ask me just to replace your AI” say yes. say you’ll owe them one but you appreciate them so much more than AI. and show up for them […]mastodon.socialOriginal post on mastodon.social 1237 37
Reposted by Leo Ré JorgeTechAltar @techaltar.mas.to.ap.brid.gy · 16/09/2026The new Apple Watch features Audio Intelligence: it constantly listens, transcribes & analyzes everything around itself, including on Apple's servers. Here's what that means for privacy Nebula: nebula.tv/videos/techaltar-the-alwa… YouTube: youtu.be/GyxYEYsNUn4 108 8
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 14/09/2026@dalias I don't get how such an obvious scam like her keeps relevant for such a long time in this space. 000 0
Reposted by Leo Ré Jorgemyrmepropagandist @futurebird.sauropods.win.ap.brid.gy · 14/09/2026Everyone knows that in robotics many tasks are easy, but seemingly simple tasks, such as cleaning a room or doing the dishes are hard. Well I have solved every room cleaning task. First the entire room is fed into an industrial shredding machine, the valuable metals are recovered by melting the […]sauropods.winOriginal post on sauropods.win 4541 41
Reposted by Leo Ré Jorgehex @hex.kolektiva.social.ap.brid.gy · 12/09/2026Let's actually talk about the "AI destroying the world" thing. One specific apocalypse scenario is the "paperclip maximizer" problem. The scenario goes that a super intelligence is tasked with maximizing the number of paperclips produced by a company. Because of a thing called "instrumental […]kolektiva.socialOriginal post on kolektiva.social 4127 27
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 09/09/2026@futurebird Brazil has exactly what you're proposing, it's called pix and it makes Trump extremely angry because it sidelines visa and MasterCard. @Techaltar has a nice video about that and other initiatives in other places m.youtube.com/watch?v=ls-eE8FVM8E 001 1
Reposted by Leo Ré Jorgejonny (nonvenomous) @jonny.neuromatch.social.ap.brid.gy · 08/09/2026OpenAI claims that they used 300 billion output tokens from a swarm of ~10,000 concurrent instances of an internal model that is almost certainly larger and more expensive to run than GPT-6. That's the total token output, including exploring related problems, which, along with spying on other […]neuromatch.socialOriginal post on neuromatch.social 001 1
Reposted by Leo Ré JorgeGeorge Monbiot @georgemonbiot.bsky.social · 07/09/202680 year-olds with pro-Palestine signs: arrested as terrorists. Greens planning to block a road, to highlight an existential crisis: sent down for years. Masked far-right thugs blocking roads and ports, using intimidation, obstruction & threatening behaviour: no arrests. The real two-tier policing. 9533571382 3.4K
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 06/09/2026@futurebird Great one! Feels like a Philip K Dick take (as others here already mentioned) on the Truman Show 000 0
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 06/09/2026[software, R, data visualization] @schluff I'm totally with you in a desire to distance from Posit though, currently looking for some replacement for Rstudio, which after 15 years is probably as big of a change 000 0
Reposted by Leo Ré JorgeGlyph @glyph.im · 03/09/2026believing that LLMs will soon be able to take on economically valuable tasks fully autonomously because their outputs sound increasingly plausible and surely it's just a matter of extending that intelligence into agency is like believing that pretty soon video games are going to have food in […]mastodon.socialOriginal post on mastodon.social 3033 33
Reposted by Leo Ré JorgeChuck Darwin @cdarwin.c.im.ap.brid.gy · 03/09/2026New York City Mayor Zohran Mamdani has unveiled a one-year ban on the use of student-facing generative AI across the city’s public schools through eighth grade, and restrictions on the use of the technology in high schools. “The tech industry wants us to believe that A.I.-powered early […]c.imOriginal post on c.im 1452 52
Reposted by Leo Ré Jorgemyrmepropagandist @futurebird.sauropods.win.ap.brid.gy · 02/09/2026Is this because it trained on these sites? Probably. But the LLM isn't designed to determine what part of the training data was significant for a particular output. Instead, after writing a response, like an undergraduate student who forgot to cite sources, it looks through the web for the […]sauropods.winOriginal post on sauropods.win 000 0
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 01/09/2026@futurebird In English D.O.I., in Portuguese doi 000 0
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 01/09/2026@futurebird the easiest analogy is ISBN for research documents 000 0
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 26/08/2026@davidgerard Do you know where this is from? There is a Brazilian song from the 80s with the lyrics: "My heroes died of overdose, my enemies are in power" 100 1
Reposted by Leo Ré Jorgemyrmepropagandist @futurebird.sauropods.win.ap.brid.gy · 25/08/2026Not everyone will be a writer, but everyone has a way of expressing their ideas that is unique and that conveys more that the bullet point summary of the paragraph. Finding the mode of expression through speaking and writing is part of the self-discovery process, it's part of growing up and […]sauropods.winOriginal post on sauropods.win 0515 15
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 25/08/2026@Techaltar No worries, I'm definitely not looking for internet points. By the way, I was very glad you also mentioned the environmental/food price issues with ethanol as fuel. 001 1
Reposted by Leo Ré JorgeTechAltar @techaltar.mas.to.ap.brid.gy · 24/08/2026Why different countries adopt different cars Nebula: nebula.tv/videos/techaltar-why-diff… 105 5
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 24/08/2026@Techaltar Really happy to have triggered your dive into this rabbit hole! The video is pretty great! 100 1
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 24/08/2026@Techaltar oh you made a video about this, very nice! I'll watch later when I'm on the computer. 100 1
Reposted by Leo Ré JorgeCassandrich @dalias.hachyderm.io.ap.brid.gy · 19/08/2026I want a Firefox extension that knows how to replace every single "N months ago" label with the actual exact timestamp (using patterns for how the data is hidden in tooltips, click-to-expand, etc.). That shit is SO unhelpful. Like if I'm looking back at git history, I don't care that a change […]hachyderm.ioOriginal post on hachyderm.io 12775 75
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 18/08/2026@davidgerard seems to be an academic politician much more than a scientist... Even working with Corona viruses, he has a handful of publications in the last 3 years, none as first or senior author. Has been vice dean of 2 different universities in that period. In that whole piece there's 0 […]mastodon.socialOriginal post on mastodon.social 001 1
Reposted by Leo Ré JorgeDavid Gerard @davidgerard.circumstances.run.ap.brid.gy · 18/08/2026Human bot defends being a human bot > AI creates almost all the final sentences in that process. That is true. But www.nature.com/articles/d44148-026-… archive archive.is/fo6xv > I remain the expert and the thought leader behind the piece. i.e. he wrote the prompt i […]circumstances.runOriginal post on circumstances.run 737 7
Reposted by Leo Ré Jorgemx alex tax1a - 2020 (6) @atax1a.infosec.exchange.ap.brid.gy · 17/08/2026cannot believe that the current debate in open source is "should we actually understand the code we write or should we give money to sam altman and dario amodei to rent the ability to produce the jankiest garbage you've ever seen" 2619 19
Reposted by Leo Ré JorgeDepths of Wiktionary @depthsofwiktionary.wikis.world.ap.brid.gy · 15/08/2026 380104 104
Reposted by Leo Ré Jorgevincentholst.bsky.social @vincentholst.bsky.social · 13/08/2026In 2023, @nature.com published 'Papers and patents are becoming less disruptive over time', receiving world-wide media attention. Our Matters Arising, published after a 32 month delay (more on that soon), shows that the reported decline can largely be attributed to dataset artefacts. 🧵 9353173 353
Reposted by Leo Ré JorgeAmbivalent Dictator For Life @pikesley.mastodon.me.uk.ap.brid.gy · 14/08/2026First they ignore you. Then they laugh at you. Then they fight you. Then you win. Then you shove a load of toxic slop into the kernel. 007 7
Reposted by Leo Ré JorgeMartin Rundkvist @mrundkvist.archaeo.social.ap.brid.gy · 05/08/2026A lot of people seem to believe that if you ask an LLM to look through a database and calculate something, or extract data according to some criteria, then it will do exactly what you ask for. It won't. It will calculate an output that looks normal. And you won't discover this unless you do the […]archaeo.socialOriginal post on archaeo.social 4528 28
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 03/08/2026@tael @zzt @davidgerard >Release Date: 9 Mar, 2021 “We expect Early Access to last about 2 weeks.” Priceless! maybe @gamingonlinux knows how to report undisclosed AI use? 011 1
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 02/08/2026@futurebird I just started last week to put up my old photos in iNaturalist and was quite impressed by the community. I didn't expect insect photos from Brazil to be so easily identified by volunteers/hobbyists, many times down to species 000 0
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 02/08/2026Damn, another bad victim of the slop flood are puzzles! I'm looking for something for my toddler, and anything that is not a licensed IP, is all very bad, obvious and cringey slop. 000 0
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 29/07/2026@0xabad1dea Thanks for the very straightforward explanation, I saw some mathematicians discussing this case, but it was all a bit over my head, this is very understandable! 010 1
Reposted by Leo Ré JorgeDavid Gerard @davidgerard.circumstances.run.ap.brid.gy · 28/07/2026has anyone documented the step by step process of how this particular swarm of parasitical wasps got into the C-level at the Wikimedia Foundation and do they even have an end game, or are they just ideologically stupid 257 7
Reposted by Leo Ré JorgeBastian Greshake Tzovaras @gedankenstuecke.scholar.social.ap.brid.gy · 22/07/2026I co-proposed/wrote the #Codeberg vibe-coding ban and all I got was a lot of haters in my mentions. And a nearly 70% approval rate, with one of the highest voter turnouts the association had afaik (and one that's comparable to the last EU elections). 😎 71343 43
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 20/07/2026@dlakelan That's a painful thread to read... and the piece itself is also pretty bad, the way the author dehumanizes the human expert to put them on the same footing as the chatbot is very baffling. I kinda follow Andrew's blog but I started just skipping anything posted by the obvious AI boosters. 001 1
Reposted by Leo Ré JorgeDaniel Lakeland @dlakelan.mastodon.sdf.org.ap.brid.gy · 20/07/2026statmodeling.stat.columbia.edu/2024… Re this comment and others on that page. Ultimately the thing that makes people talk about AI as "slop" and distrust people who use it is that its impossible to gain the speed advantages […]mastodon.sdf.orgOriginal post on mastodon.sdf.org 102 2
Reposted by Leo Ré Jorgehex @hex.kolektiva.social.ap.brid.gy · 18/07/2026I keep seeing the "cars replaced horses" thing related to "AI." People keep comparing "AI" to cars (positively) without thinking about the fact that cars were forced on people (AAA bought up and destroyed tram lines, there were massive protests, etc), and are the primary driver of climate […]kolektiva.socialOriginal post on kolektiva.social 818170 170
Leo Ré Jorge @leorjorge.mastodon.social.ap.brid.gy · 07/07/2026Looking for #bookstodon tips! I'm father of a very curious 3.5 year old, who is excited about maps and planets. Most atlases I see aimed at kids are full of bad stereotyping, so any #map nerds with kids and/or suggestions, I'm on the lookout. Similarly, any #space enthusiasts (maybe @badastro?) […]mastodon.socialOriginal post on mastodon.social 021 2
Reposted by Leo Ré JorgeGlyph @glyph.im · 23/06/2026The thing that “AIs” are the best at is turning every communication into a conflict, where the LLM user is the attacker and the person who has to consume its output is the defender. blog.glyph.im/2026/06/adversarial-c…blog.glyph.imDeciphering Glyph :: Adversarial CommunicationDeciphering Glyph, the blog of Glyph Lefkowitz. 5751 51
