Sergey Ast
@sergey-astakhov.bsky.social
Social researcher of emerging technologies
Reposted by Sergey Ast
Promote people of colour and women in this space, whistleblowers and survivors of these clowns.
Read more on actual critical concepts here:
bsky.app/profile/oliv...
Read more on actual critical concepts here:
bsky.app/profile/oliv...
CAIL = prerequisite knowledge for a critical perspective, such as to tell apart nonsense hype from true theoretical computer scientific claims. For example, the idea that human-like systems are a sensible or possible goal is the result of circular reasoning and anthropomorphism. olivia.science/ai
🧵
🧵
November 30, 2025 at 7:09 AM
Promote people of colour and women in this space, whistleblowers and survivors of these clowns.
Read more on actual critical concepts here:
bsky.app/profile/oliv...
Read more on actual critical concepts here:
bsky.app/profile/oliv...
From the wonderful @andreskarjus.bsky.social paper: "Humans too are stochastic black boxes. Engaging in analytic tasks requiring subjective judgments and reasoning can propagate and amplify biases.
November 28, 2025 at 4:32 PM
From the wonderful @andreskarjus.bsky.social paper: "Humans too are stochastic black boxes. Engaging in analytic tasks requiring subjective judgments and reasoning can propagate and amplify biases.
Reposted by Sergey Ast
Sometimes people, like programs, get caught in a loop.
These loops can last minutes, these loops can last years.
But the solution is the same.
Just turn it off and on again.
These loops can last minutes, these loops can last years.
But the solution is the same.
Just turn it off and on again.
November 27, 2025 at 2:32 PM
Sometimes people, like programs, get caught in a loop.
These loops can last minutes, these loops can last years.
But the solution is the same.
Just turn it off and on again.
These loops can last minutes, these loops can last years.
But the solution is the same.
Just turn it off and on again.
I wonder whether creative communities would be fine with ChatGPT, Midjourney, DALL·E, and similar tools if the major IP issues were somehow resolved.
November 27, 2025 at 1:02 PM
I wonder whether creative communities would be fine with ChatGPT, Midjourney, DALL·E, and similar tools if the major IP issues were somehow resolved.
Reposted by Sergey Ast
My new book, The Division of Rationalized Labor, is now shipping! A brief summary of the argument to follow…
November 26, 2025 at 5:45 PM
My new book, The Division of Rationalized Labor, is now shipping! A brief summary of the argument to follow…
The easiest way to measure #procrastination is to track how many irrelevant tasks you’ve accomplished as the deadline approaches.
November 26, 2025 at 5:23 PM
The easiest way to measure #procrastination is to track how many irrelevant tasks you’ve accomplished as the deadline approaches.
Reposted by Sergey Ast
Reposted by Sergey Ast
Qualitative research with LLM chatbots: Technological reflexivity for interpretative technology
journals.sagepub.com/doi/full/10....
I haven't tried it yet, but I think LLMs can be useful for some tasks for some kind of qualitative research. wrt the article, I am not sure how consistent the 1/
journals.sagepub.com/doi/full/10....
I haven't tried it yet, but I think LLMs can be useful for some tasks for some kind of qualitative research. wrt the article, I am not sure how consistent the 1/
November 26, 2025 at 11:38 AM
Qualitative research with LLM chatbots: Technological reflexivity for interpretative technology
journals.sagepub.com/doi/full/10....
I haven't tried it yet, but I think LLMs can be useful for some tasks for some kind of qualitative research. wrt the article, I am not sure how consistent the 1/
journals.sagepub.com/doi/full/10....
I haven't tried it yet, but I think LLMs can be useful for some tasks for some kind of qualitative research. wrt the article, I am not sure how consistent the 1/
Just a reminder to myself and everyone: writing can suck even without ChatGPT.
November 26, 2025 at 9:47 AM
Just a reminder to myself and everyone: writing can suck even without ChatGPT.
Reposted by Sergey Ast
We have a new paper in Science Advances proposing a simple test for bias:
Is the same person treated differently when their race is perceived differently?
Specifically, we study: is the same driver likelier to be searched by police when they are perceived as Hispanic rather than white?
1/
Is the same person treated differently when their race is perceived differently?
Specifically, we study: is the same driver likelier to be searched by police when they are perceived as Hispanic rather than white?
1/
November 24, 2025 at 6:14 PM
We have a new paper in Science Advances proposing a simple test for bias:
Is the same person treated differently when their race is perceived differently?
Specifically, we study: is the same driver likelier to be searched by police when they are perceived as Hispanic rather than white?
1/
Is the same person treated differently when their race is perceived differently?
Specifically, we study: is the same driver likelier to be searched by police when they are perceived as Hispanic rather than white?
1/
The strangest thing about hype as a phenomenon is that, decade after decade, we continue to be annoyed by the same, almost natural cycles of public attention. Obviously, that annoyance is part of the hype itself.
November 24, 2025 at 9:24 AM
The strangest thing about hype as a phenomenon is that, decade after decade, we continue to be annoyed by the same, almost natural cycles of public attention. Obviously, that annoyance is part of the hype itself.
Reposted by Sergey Ast
One of the themes of HOW INFRASTRUCTURE WORKS is that infrastructural systems physically manifest a relationship between people and with the place they live that extends into the future. In nations it’s often by fiat, but in cities it’s because of network effects and because water runs downhill.
Banal observation, sorry, but conversations in Prague really brought home to me that while countries come and go, cities have a historical existence and character that typically is far more resilient.
November 14, 2025 at 11:35 AM
One of the themes of HOW INFRASTRUCTURE WORKS is that infrastructural systems physically manifest a relationship between people and with the place they live that extends into the future. In nations it’s often by fiat, but in cities it’s because of network effects and because water runs downhill.
Can't believe I remember the time when AI was a small field for specialists.
November 21, 2025 at 6:36 PM
Can't believe I remember the time when AI was a small field for specialists.
When it comes to writing, I wouldn’t pretend that LLMs don’t exist. But I wouldn’t rely on them exclusively, either. I’d rather hone my skills by interleaving periods of using ChatGPT with periods of old-school, organic writing.
November 20, 2025 at 2:56 PM
When it comes to writing, I wouldn’t pretend that LLMs don’t exist. But I wouldn’t rely on them exclusively, either. I’d rather hone my skills by interleaving periods of using ChatGPT with periods of old-school, organic writing.
Reposted by Sergey Ast
A semi-regular AI map-drawing test. This was my older attempt
November 20, 2025 at 10:07 AM
A semi-regular AI map-drawing test. This was my older attempt
I read a lot, but every time I even touch a book or a story, I feel this weird resistance. It never fully goes away, so I keep checking how many pages are left.
November 19, 2025 at 7:16 PM
I read a lot, but every time I even touch a book or a story, I feel this weird resistance. It never fully goes away, so I keep checking how many pages are left.
Reposted by Sergey Ast
I’ve written a piece on the curious lack of media and political interest in the issues faced by our national @britishlibrary.bsky.social. This is strange given we live in a world where ideas, knowledge and research are a long-term source of innovation and insight
www.cityam.com/the-british-...
www.cityam.com/the-british-...
The British library is in crisis: why does nobody care?
The widespread indifference to the British Library's crippling cyberattack demonstrates a perilous failure to value the knowledge infrastructure vital for national prosperity
www.cityam.com
November 18, 2025 at 6:27 AM
I’ve written a piece on the curious lack of media and political interest in the issues faced by our national @britishlibrary.bsky.social. This is strange given we live in a world where ideas, knowledge and research are a long-term source of innovation and insight
www.cityam.com/the-british-...
www.cityam.com/the-british-...
There must be a deep connection between the benefit of the doubt and the suspension of disbelief.
November 14, 2025 at 12:19 AM
There must be a deep connection between the benefit of the doubt and the suspension of disbelief.
Did you know that cocoa and cacao are two different things?
November 13, 2025 at 5:15 PM
Did you know that cocoa and cacao are two different things?
Reposted by Sergey Ast
What is the most profitable industry in the world, this side of the law? Not oil, not IT, not pharma.
It's *scientific publishing*.
We call this the Drain of Scientific Publishing.
Paper: arxiv.org/abs/2511.04820
Background: doi.org/10.1162/qss_...
Thread @markhanson.fediscience.org.ap.brid.gy 👇
It's *scientific publishing*.
We call this the Drain of Scientific Publishing.
Paper: arxiv.org/abs/2511.04820
Background: doi.org/10.1162/qss_...
Thread @markhanson.fediscience.org.ap.brid.gy 👇
November 12, 2025 at 10:31 AM
What is the most profitable industry in the world, this side of the law? Not oil, not IT, not pharma.
It's *scientific publishing*.
We call this the Drain of Scientific Publishing.
Paper: arxiv.org/abs/2511.04820
Background: doi.org/10.1162/qss_...
Thread @markhanson.fediscience.org.ap.brid.gy 👇
It's *scientific publishing*.
We call this the Drain of Scientific Publishing.
Paper: arxiv.org/abs/2511.04820
Background: doi.org/10.1162/qss_...
Thread @markhanson.fediscience.org.ap.brid.gy 👇
True maturity means separating not only from your parents, but also from your government. After all, nation-states have countless ways of making you work against your own interests.
November 12, 2025 at 9:02 AM
True maturity means separating not only from your parents, but also from your government. After all, nation-states have countless ways of making you work against your own interests.
Every time I hear something about social bubbles, my first reaction is: social bubbles are natural.
November 10, 2025 at 7:11 PM
Every time I hear something about social bubbles, my first reaction is: social bubbles are natural.
There’s a particular kind of pleasure in experiencing art that isn’t perfect.
November 10, 2025 at 8:52 AM
There’s a particular kind of pleasure in experiencing art that isn’t perfect.
Reposted by Sergey Ast
Today feels like a good day to remind everyone that there is literally no empirical evidence that economic growth will just continue if the world heats up to, say, 3°C by 2100.
Continued growth is just *assumed* in every economic model of the relationship between temperatures and growth.
🧵
Continued growth is just *assumed* in every economic model of the relationship between temperatures and growth.
🧵
November 6, 2025 at 1:52 PM
Today feels like a good day to remind everyone that there is literally no empirical evidence that economic growth will just continue if the world heats up to, say, 3°C by 2100.
Continued growth is just *assumed* in every economic model of the relationship between temperatures and growth.
🧵
Continued growth is just *assumed* in every economic model of the relationship between temperatures and growth.
🧵
Reposted by Sergey Ast
Slides from my @mit.edu IDSS Distinguished Speaker seminar "Networks untangle gender differences in productivity and prominence among scientists" this week
I argue that collaboration networks act like unequally distributed (and gendered) social capital
aaronclauset.github.io/slides/Claus...
I argue that collaboration networks act like unequally distributed (and gendered) social capital
aaronclauset.github.io/slides/Claus...
November 6, 2025 at 6:05 PM
Slides from my @mit.edu IDSS Distinguished Speaker seminar "Networks untangle gender differences in productivity and prominence among scientists" this week
I argue that collaboration networks act like unequally distributed (and gendered) social capital
aaronclauset.github.io/slides/Claus...
I argue that collaboration networks act like unequally distributed (and gendered) social capital
aaronclauset.github.io/slides/Claus...