aster
@aqperkins.bsky.social
(they/them). Ph.D. in Neuroscience. Views my own.
Reposted by aster
"there is little reason to doubt what the aim [of the massive investment in AI] is: to embed knowledge in machines—knowledge that previously belonged to the people working in these domains."
🧵
🧵
In applying AI to material science, biology, etc, capitalism is trying to shed science.
The point is to substitute the engineering of a machine that can generate what science has hitherto done, but without having people know things. Knowledge ultimately residing in private property is the dream.
The point is to substitute the engineering of a machine that can generate what science has hitherto done, but without having people know things. Knowledge ultimately residing in private property is the dream.
Jeff Bezos Creates A.I. Start-Up Where He Will Be Co-Chief Executive
www.nytimes.com
December 10, 2025 at 6:20 PM
"there is little reason to doubt what the aim [of the massive investment in AI] is: to embed knowledge in machines—knowledge that previously belonged to the people working in these domains."
🧵
🧵
Reposted by aster
A week after the arXiv was forced to tighten down on submissions because of overwhelming volumes of AI slop, bioRxiv is throwing the doors wide open.
disappointed that @biorxivpreprint.bsky.social is implicitly endorsing the use of LLMs to replace scientific thought
@richardsever.bsky.social this is a short-sighted move and a net negative for science
@richardsever.bsky.social this is a short-sighted move and a net negative for science
IT'S HAPPENING! 💥 I'm psyched to launch the collaboration between @qedscience.bsky.social & @openrxiv.bsky.social @biorxivpreprint.bsky.social! Preprint + q.e.d = your science is out there, and anyone can appreciate it. Let's care about making discoveries, and not on “getting published” (1/3) 👇
November 7, 2025 at 1:59 PM
A week after the arXiv was forced to tighten down on submissions because of overwhelming volumes of AI slop, bioRxiv is throwing the doors wide open.
Reposted by aster
Undetstanding well your data is a key part of your analysis. It takes time. There are already too many papers out with poorly understood data that was run through a statistical model just to get a significant result. Adding llms to mix is just adding fuel to the fire.
the amount of academic research now using llms in research (synthetic data generation, to classify, annotate, or analyse large scale data, etc) is astounding. remember, just cus use of llms in research is becoming normalised does NOT erase the fact it degrades the research & undermine your results
October 29, 2025 at 6:50 PM
Undetstanding well your data is a key part of your analysis. It takes time. There are already too many papers out with poorly understood data that was run through a statistical model just to get a significant result. Adding llms to mix is just adding fuel to the fire.
Reposted by aster
That U.S. academics didn't collectively develop strong labor consciousness is really going to bite them in the ass as universities bend the knee to this administration. Politely worded letters to presidents and trustees are not going to count for much of anything.
October 23, 2025 at 5:12 PM
That U.S. academics didn't collectively develop strong labor consciousness is really going to bite them in the ass as universities bend the knee to this administration. Politely worded letters to presidents and trustees are not going to count for much of anything.
Reposted by aster
we have the clarity to say it's obscene that a 21 year old named "big ballz" challenges government employees to prove their work isn't bullshit that a chatbot can do, but we apparently don't have the clarity to say the same when it's geoffrey hinton or bill gates talking about healthcare & education
August 12, 2025 at 11:39 PM
we have the clarity to say it's obscene that a 21 year old named "big ballz" challenges government employees to prove their work isn't bullshit that a chatbot can do, but we apparently don't have the clarity to say the same when it's geoffrey hinton or bill gates talking about healthcare & education
Reposted by aster
Happy to have some good science news - our new study with @aqperkins.bsky.social is out in @plosbiology.org!
dx.plos.org/10.1371/jour...
dx.plos.org/10.1371/jour...
Orbitofrontal cortex computes gaze-dependent comparisons between attributes rather than integrated values
When choosing between options with multiple attributes, the orbitofrontal cortex (OFC) is thought to integrate across those attributes, aiding in value comparisons. This study shows in monkeys that OF...
dx.plos.org
August 7, 2025 at 8:13 PM
Happy to have some good science news - our new study with @aqperkins.bsky.social is out in @plosbiology.org!
dx.plos.org/10.1371/jour...
dx.plos.org/10.1371/jour...
Reposted by aster
"Asked to generate intervention plans for struggling students, AI teacher assistants recommended more-punitive measures for hypothetical students with Black-coded names and more supportive approaches for students the platforms perceived as white" www.chalkbeat.org/2025/08/06/a...
Annie and Lakeesha struggle in school. AI teacher assistants treated them very differently.
A Common Sense Media study found that prominent teacher assistants that use AI generated recommendations that appeared to be rooted in racial stereotypes based on students’ names. About a third of tea...
www.chalkbeat.org
August 6, 2025 at 7:25 PM
"Asked to generate intervention plans for struggling students, AI teacher assistants recommended more-punitive measures for hypothetical students with Black-coded names and more supportive approaches for students the platforms perceived as white" www.chalkbeat.org/2025/08/06/a...
Reposted by aster
and yet, we'll use behavioural and cognitive sciences language to mislead people into thinking AI models equate human cognition
A very odd thing about Artificial Intelligence as a discipline in computer science is that it historically shifted from “understanding the human brain better” to “we give up on understanding the brain and will just replace humans despite having no fucking clue”
July 7, 2025 at 3:41 PM
and yet, we'll use behavioural and cognitive sciences language to mislead people into thinking AI models equate human cognition
Reposted by aster
New paper hot off the press www.nature.com/articles/s41...
We analysed over 40,000 computer vision papers from CVPR (the longest standing CV conf) & associated patents tracing pathways from research to application. We found that 90% of papers & 86% of downstream patents power surveillance
1/
We analysed over 40,000 computer vision papers from CVPR (the longest standing CV conf) & associated patents tracing pathways from research to application. We found that 90% of papers & 86% of downstream patents power surveillance
1/
Computer-vision research powers surveillance technology - Nature
An analysis of research papers and citing patents indicates the extensive ties between computer-vision research and surveillance.
www.nature.com
June 25, 2025 at 5:29 PM
New paper hot off the press www.nature.com/articles/s41...
We analysed over 40,000 computer vision papers from CVPR (the longest standing CV conf) & associated patents tracing pathways from research to application. We found that 90% of papers & 86% of downstream patents power surveillance
1/
We analysed over 40,000 computer vision papers from CVPR (the longest standing CV conf) & associated patents tracing pathways from research to application. We found that 90% of papers & 86% of downstream patents power surveillance
1/
Reposted by aster
Teaching "A Critical Introduction to AI Images" this semester as a visiting prof at RIT's Humanities, Computing and Design department. Here's the reading list I have so far.
January 9, 2025 at 8:36 PM
Teaching "A Critical Introduction to AI Images" this semester as a visiting prof at RIT's Humanities, Computing and Design department. Here's the reading list I have so far.
Reposted by aster
i genuinely try to be open about the usefulness and advantages of llms and try entertain ideas I deeply disagree with. and there’s a current trend that attempts to occupy a middle-ground position where people claims to find useful use cases while at the same time claiming to be critical
1/
1/
January 1, 2025 at 6:30 PM
i genuinely try to be open about the usefulness and advantages of llms and try entertain ideas I deeply disagree with. and there’s a current trend that attempts to occupy a middle-ground position where people claims to find useful use cases while at the same time claiming to be critical
1/
1/
Reposted by aster
In 2010, anti-copyright activist Aaron Swartz hanged himself after being prosecuted to the full extent of the law for making copyrighted academic publications freely available. Today, academic publishers are insisting those copyrights be lifted to feed the corporate AI slop machine.
‘Impossible’ to create AI tools like ChatGPT without copyrighted material, OpenAI says
‘Impossible’ to create AI tools like ChatGPT without copyrighted material, OpenAI says
Pressure grows on artificial intelligence firms over the content used to train their productsBusiness live – latest updatesThe developer OpenAI has said it would be impossible to create tools like its groundbreaking chatbot ChatGPT without access to copyrighted material, as pressure grows on artificial intelligence firms over the content used to train their products.Chatbots such as ChatGPT and image generators like Stable Diffusion are “trained” on a vast trove of data taken from the internet, with much of it covered by copyright – a legal protection against someone’s work being used without permission. Continue reading...
www.theguardian.com
December 18, 2024 at 1:25 PM
In 2010, anti-copyright activist Aaron Swartz hanged himself after being prosecuted to the full extent of the law for making copyrighted academic publications freely available. Today, academic publishers are insisting those copyrights be lifted to feed the corporate AI slop machine.
Isoflurane is a beautiful name for a baby girl
December 6, 2024 at 8:03 PM
Isoflurane is a beautiful name for a baby girl