@manrv.bsky.social
Reposted
New preprint! @marentierra.bsky.social @irisvanrooij.bsky.social & I have been working on what CAIL means to showcase & propagate the idea of thinking very differently to tech industry norms on "artificial intelligence"
Towards Critical Artificial Intelligence Literacies doi.org/10.5281/zeno...
1/
Towards Critical Artificial Intelligence Literacies doi.org/10.5281/zeno...
1/
December 2, 2025 at 6:58 AM
New preprint! @marentierra.bsky.social @irisvanrooij.bsky.social & I have been working on what CAIL means to showcase & propagate the idea of thinking very differently to tech industry norms on "artificial intelligence"
Towards Critical Artificial Intelligence Literacies doi.org/10.5281/zeno...
1/
Towards Critical Artificial Intelligence Literacies doi.org/10.5281/zeno...
1/
Reposted
This is terrifying.
"[AI agents] can... infer a researcher's latent hypotheses and produce data that artificially confirms them."
...
"We can no longer trust that survey responses are coming from real people" [email protected]
"[AI agents] can... infer a researcher's latent hypotheses and produce data that artificially confirms them."
...
"We can no longer trust that survey responses are coming from real people" [email protected]
November 18, 2025 at 9:03 PM
This is terrifying.
"[AI agents] can... infer a researcher's latent hypotheses and produce data that artificially confirms them."
...
"We can no longer trust that survey responses are coming from real people" [email protected]
"[AI agents] can... infer a researcher's latent hypotheses and produce data that artificially confirms them."
...
"We can no longer trust that survey responses are coming from real people" [email protected]
Reposted
So why do I let AI in my house? Don’t I care about the environmental consequences? What about the theft of intellectual property? Don’t I understand that these are bullshit machines?
Yes, yes, yes, and yes.
It’s what I do all day long.
Yes, yes, yes, and yes.
It’s what I do all day long.
Modern-Day Oracles or Bullshit Machines: Introduction
A free online humanities course about how to learn and work and thrive in an AI world.
thebullshitmachines.com
October 18, 2025 at 6:26 PM
So why do I let AI in my house? Don’t I care about the environmental consequences? What about the theft of intellectual property? Don’t I understand that these are bullshit machines?
Yes, yes, yes, and yes.
It’s what I do all day long.
Yes, yes, yes, and yes.
It’s what I do all day long.
Reposted
Boiling here at home in Cyprus but I put the finishing touches a couple of days ago on this preprint: What Does 'Human-Centred AI' Mean? doi.org/10.48550/arX...
Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n
Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n
July 29, 2025 at 11:52 AM
Boiling here at home in Cyprus but I put the finishing touches a couple of days ago on this preprint: What Does 'Human-Centred AI' Mean? doi.org/10.48550/arX...
Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n
Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n
Valuable point of view and academic reflection
AI slop and the destruction of knowledge irisvanrooijcogsci.com/2025/08/12/a...
AI slop and the destruction of knowledge
This week I was looking for info on what cognitive scientists mean when they speak of ‘domain-general’ cognition. I was curious, because the nuances are relevant for something I am researching at t…
irisvanrooijcogsci.com
August 14, 2025 at 12:15 PM
Valuable point of view and academic reflection
Reposted
AI slop and the destruction of knowledge irisvanrooijcogsci.com/2025/08/12/a...
AI slop and the destruction of knowledge
This week I was looking for info on what cognitive scientists mean when they speak of ‘domain-general’ cognition. I was curious, because the nuances are relevant for something I am researching at t…
irisvanrooijcogsci.com
August 12, 2025 at 10:12 PM
AI slop and the destruction of knowledge irisvanrooijcogsci.com/2025/08/12/a...
Reposted
The current state of AI for sustained work: exponential progress continues with no unexpected leaps but also no walls.
(The METR measure is just one of many benchmarks, and like all benchmarks has flaws, but also has the advantage of have neither a ceiling or floor effect)
metr.org/blog/2025-03...
(The METR measure is just one of many benchmarks, and like all benchmarks has flaws, but also has the advantage of have neither a ceiling or floor effect)
metr.org/blog/2025-03...
August 9, 2025 at 2:00 PM
The current state of AI for sustained work: exponential progress continues with no unexpected leaps but also no walls.
(The METR measure is just one of many benchmarks, and like all benchmarks has flaws, but also has the advantage of have neither a ceiling or floor effect)
metr.org/blog/2025-03...
(The METR measure is just one of many benchmarks, and like all benchmarks has flaws, but also has the advantage of have neither a ceiling or floor effect)
metr.org/blog/2025-03...
Reposted
New prompting report, from us: Don't bother threatening your AI.
Does threatening an AI really make it perform better (the way Google founder Brin claimed)? How about offering to tip the AI? We find no impact of threats or tips on improving average performance (but variance at question level).
Does threatening an AI really make it perform better (the way Google founder Brin claimed)? How about offering to tip the AI? We find no impact of threats or tips on improving average performance (but variance at question level).
August 1, 2025 at 2:59 PM
New prompting report, from us: Don't bother threatening your AI.
Does threatening an AI really make it perform better (the way Google founder Brin claimed)? How about offering to tip the AI? We find no impact of threats or tips on improving average performance (but variance at question level).
Does threatening an AI really make it perform better (the way Google founder Brin claimed)? How about offering to tip the AI? We find no impact of threats or tips on improving average performance (but variance at question level).
Reposted
Las interacciones con máquinas pueden modificar nuestras expectativas sobre las relaciones humanas al habituarnos a interacciones rápidas, perfectas y sin conflicto. Cuando el proceso comunicativo NUNCA es así. De eso hablé en #LasQueCuentanLaCiencia 👉 el efecto ELIZA
www.youtube.com/watch?v=sVhP...
www.youtube.com/watch?v=sVhP...
Las que cuentan la ciencia 2025 - Lorena Fernández 'El efecto ELIZA'
YouTube video by Córdoba Ciencia
www.youtube.com
June 26, 2025 at 8:00 PM
Las interacciones con máquinas pueden modificar nuestras expectativas sobre las relaciones humanas al habituarnos a interacciones rápidas, perfectas y sin conflicto. Cuando el proceso comunicativo NUNCA es así. De eso hablé en #LasQueCuentanLaCiencia 👉 el efecto ELIZA
www.youtube.com/watch?v=sVhP...
www.youtube.com/watch?v=sVhP...
Reposted
Just FYI. Qwen3-Embedding models are number 2, 3, and 4 in MTEB Leaderboard.
huggingface.co/spaces/mteb/...
huggingface.co/spaces/mteb/...
June 7, 2025 at 1:00 PM
Just FYI. Qwen3-Embedding models are number 2, 3, and 4 in MTEB Leaderboard.
huggingface.co/spaces/mteb/...
huggingface.co/spaces/mteb/...
Reposted
🧪 a really interesting study showing that people respond differently to types of carbohydrates. “…our results demonstrate that different strategies may be useful for nutrient intervention in groups with different ethnicities and metabolic functions.”
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Individual variations in glycemic responses to carbohydrates and underlying metabolic physiology - Nature Medicine
Deep phenotyping of responses to carbohydrate meals and mitigators revealed interindividual differences in postprandial glycemic responses that reflect underlying metabolic physiology, such as insulin...
www.nature.com
June 6, 2025 at 8:30 PM
🧪 a really interesting study showing that people respond differently to types of carbohydrates. “…our results demonstrate that different strategies may be useful for nutrient intervention in groups with different ethnicities and metabolic functions.”
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reposted
New paper just dropped🎉 With novel "Curiosity Boxes", we find that chimps & children are very curious about social interactions, & some even give up a reward to gain info! Fun collaboration with @alisongopnik.bsky.social, @janengelmann.bsky.social & others royalsocietypublishing.org/doi/10.1098/...
Chimpanzees and children are curious about social interactions | Proceedings of the Royal Society B: Biological Sciences
Curiosity is adaptive, enhances learning, and reduces uncertainty. Social curiosity
is defined as the motivation to gain information about the actions, relationships,
and psychology of others. Little ...
royalsocietypublishing.org
June 6, 2025 at 4:44 PM
New paper just dropped🎉 With novel "Curiosity Boxes", we find that chimps & children are very curious about social interactions, & some even give up a reward to gain info! Fun collaboration with @alisongopnik.bsky.social, @janengelmann.bsky.social & others royalsocietypublishing.org/doi/10.1098/...
Reposted
Example of why I think current LLMs are enough to change lots of work even if they don’t get better, once we start integrating them with other systems
GPT-4 (now obsolete) went from 30% accuracy to 87% accuracy in clinical oncology decisions when given access to tools www.nature.com/articles/s43...
GPT-4 (now obsolete) went from 30% accuracy to 87% accuracy in clinical oncology decisions when given access to tools www.nature.com/articles/s43...
June 6, 2025 at 4:14 PM
Example of why I think current LLMs are enough to change lots of work even if they don’t get better, once we start integrating them with other systems
GPT-4 (now obsolete) went from 30% accuracy to 87% accuracy in clinical oncology decisions when given access to tools www.nature.com/articles/s43...
GPT-4 (now obsolete) went from 30% accuracy to 87% accuracy in clinical oncology decisions when given access to tools www.nature.com/articles/s43...
Reposted
There's a lot happening for predicting and preventing Alzheimer's disease. Here's the latest.
erictopol.substack.com/p/predicting...
erictopol.substack.com/p/predicting...
May 31, 2025 at 6:09 PM
There's a lot happening for predicting and preventing Alzheimer's disease. Here's the latest.
erictopol.substack.com/p/predicting...
erictopol.substack.com/p/predicting...
Reposted
Una investigación re-evalúa el papel de las mujeres en el salto tecnológico que se produjo en la Alta Edad Media www.muyinteresante.com/historia/inv...
Reivindica la agencia femenina como un factor central en la difusión y adaptación de tecnologías clave durante esa época.
Reivindica la agencia femenina como un factor central en la difusión y adaptación de tecnologías clave durante esa época.
Una innovadora investigación reevalúa el papel de las mujeres en el salto tecnológico que se produjo en la Alta Edad Media
El trabajo de las mujeres incluenció los avances técnicos de la Alta Edad Media. Una nueva investigación explora su impacto en la tecnología medieval.
www.muyinteresante.com
May 18, 2025 at 8:10 PM
Una investigación re-evalúa el papel de las mujeres en el salto tecnológico que se produjo en la Alta Edad Media www.muyinteresante.com/historia/inv...
Reivindica la agencia femenina como un factor central en la difusión y adaptación de tecnologías clave durante esa época.
Reivindica la agencia femenina como un factor central en la difusión y adaptación de tecnologías clave durante esa época.
Reposted
It is interesting to watch us human beings create our biggest problems. Why do we do it?
#AIEthics #cybersecurity
www.technologyreview.com/2025/04/04/1...
#AIEthics #cybersecurity
www.technologyreview.com/2025/04/04/1...
Cyberattacks by AI agents are coming
Agents could make it easier and cheaper for criminals to hack systems at scale. We need to be ready.
www.technologyreview.com
May 4, 2025 at 8:09 AM
It is interesting to watch us human beings create our biggest problems. Why do we do it?
#AIEthics #cybersecurity
www.technologyreview.com/2025/04/04/1...
#AIEthics #cybersecurity
www.technologyreview.com/2025/04/04/1...
Reposted
New work on multimodal framing! 💫
Some fun results: comparisons of the same frame when expressed in images vs texts. When the "crime" frame is expressed in the article text, there are more political words in the text, but when the frame is expressed in the article image, more police words.
Some fun results: comparisons of the same frame when expressed in images vs texts. When the "crime" frame is expressed in the article text, there are more political words in the text, but when the frame is expressed in the article image, more police words.
April 7, 2025 at 9:48 AM
New work on multimodal framing! 💫
Some fun results: comparisons of the same frame when expressed in images vs texts. When the "crime" frame is expressed in the article text, there are more political words in the text, but when the frame is expressed in the article image, more police words.
Some fun results: comparisons of the same frame when expressed in images vs texts. When the "crime" frame is expressed in the article text, there are more political words in the text, but when the frame is expressed in the article image, more police words.
Open Source is one of the greatest contributions to the society.
March 17, 2025 at 11:50 PM
Open Source is one of the greatest contributions to the society.
Reposted
We've known social disadvantage was linked to adverse health outcomes. A @NatureMedicine study used high- throughput proteomics to show there is accelerated aging, increased risk of 66 age-related diseases, and unfavorable changes in the immune system, ⬆️ inflammation
www.nature.com/articles/s41...
www.nature.com/articles/s41...
March 14, 2025 at 9:25 PM
We've known social disadvantage was linked to adverse health outcomes. A @NatureMedicine study used high- throughput proteomics to show there is accelerated aging, increased risk of 66 age-related diseases, and unfavorable changes in the immune system, ⬆️ inflammation
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reposted
A very exciting day for open-source AI! We're releasing our biggest open source model yet -- OLMo 2 32B -- and it beats the latest GPT 3.5, GPT 4o mini, and leading open weight models like Qwen and Mistral. As usual, all data, weights, code, etc. are available.
March 13, 2025 at 6:16 PM
A very exciting day for open-source AI! We're releasing our biggest open source model yet -- OLMo 2 32B -- and it beats the latest GPT 3.5, GPT 4o mini, and leading open weight models like Qwen and Mistral. As usual, all data, weights, code, etc. are available.
Reposted
Tired but happy to say this is out w @andreaeyleen.bsky.social: Are Neurocognitive Representations 'Small Cakes'? philsci-archive.pitt.edu/24834/
We analyse cog neuro theories showing how vicious regress, e.g. the homunculus fallacy, is (sadly) alive and well — and importantly how to avoid it. 1/
We analyse cog neuro theories showing how vicious regress, e.g. the homunculus fallacy, is (sadly) alive and well — and importantly how to avoid it. 1/
March 1, 2025 at 2:16 PM
Tired but happy to say this is out w @andreaeyleen.bsky.social: Are Neurocognitive Representations 'Small Cakes'? philsci-archive.pitt.edu/24834/
We analyse cog neuro theories showing how vicious regress, e.g. the homunculus fallacy, is (sadly) alive and well — and importantly how to avoid it. 1/
We analyse cog neuro theories showing how vicious regress, e.g. the homunculus fallacy, is (sadly) alive and well — and importantly how to avoid it. 1/
On the use of LLM's, the evidence tells us:
“Our findings reveal widespread adoption of large language models across diverse writing domains, ranging consumers, firms and international organizations.”
“Early adopters may have already reached a saturation point”
arxiv.org/abs/2502.09747
“Our findings reveal widespread adoption of large language models across diverse writing domains, ranging consumers, firms and international organizations.”
“Early adopters may have already reached a saturation point”
arxiv.org/abs/2502.09747
The past 18 months have seen the most rapid change in human written communication ever
By. September 2024, 18% of financial consumer complaints, 24% of press releases, 15% of job postings & 14% of UN press releases showed signs of LLM writing. And the method undercounts true use.
By. September 2024, 18% of financial consumer complaints, 24% of press releases, 15% of job postings & 14% of UN press releases showed signs of LLM writing. And the method undercounts true use.
March 1, 2025 at 4:47 PM
On the use of LLM's, the evidence tells us:
“Our findings reveal widespread adoption of large language models across diverse writing domains, ranging consumers, firms and international organizations.”
“Early adopters may have already reached a saturation point”
arxiv.org/abs/2502.09747
“Our findings reveal widespread adoption of large language models across diverse writing domains, ranging consumers, firms and international organizations.”
“Early adopters may have already reached a saturation point”
arxiv.org/abs/2502.09747
Reposted
📄 A set of provocations for considering the uses, impact, and harms of Gen AI from the perspective of humanities researchers. @laurenfklein.bsky.social, @mmvty.bsky.social, André Brock, @mariaa.bsky.social, @mellymeldubs.bsky.social, @jmjafrx.bsky.social, @nolauren.bsky.social, @dmimno.bsky.social
March 1, 2025 at 12:04 AM
📄 A set of provocations for considering the uses, impact, and harms of Gen AI from the perspective of humanities researchers. @laurenfklein.bsky.social, @mmvty.bsky.social, André Brock, @mariaa.bsky.social, @mellymeldubs.bsky.social, @jmjafrx.bsky.social, @nolauren.bsky.social, @dmimno.bsky.social
Reposted
Remote work doesn't thwart productivity. It boosts focus.
Government workers are 12% more productive when randomly assigned to work from home. They're more efficient where it's quiet.
Most people aren't shirking from home. They're escaping distractions and long commutes.
Government workers are 12% more productive when randomly assigned to work from home. They're more efficient where it's quiet.
Most people aren't shirking from home. They're escaping distractions and long commutes.
February 23, 2025 at 4:20 PM
Remote work doesn't thwart productivity. It boosts focus.
Government workers are 12% more productive when randomly assigned to work from home. They're more efficient where it's quiet.
Most people aren't shirking from home. They're escaping distractions and long commutes.
Government workers are 12% more productive when randomly assigned to work from home. They're more efficient where it's quiet.
Most people aren't shirking from home. They're escaping distractions and long commutes.
Wow “LECs actively regulate anti-tumor immunity via their ability to present tumor antigens”
For decades, we underestimated the importance of our lymphatic vessels to suppress cancer and its spread. Here's a new, open-access, outstanding review
rupress.org/jem/article/...
rupress.org/jem/article/...
February 23, 2025 at 7:52 PM
Wow “LECs actively regulate anti-tumor immunity via their ability to present tumor antigens”