beth semel
@bethmsemel.bsky.social
AI killjoy, prednisone hater, writing a book about voice analysis AI and the limits of technosolutionist listening in American mental healthcare
Reposted by beth semel
"everyone is talking about machine learning but no one is talking about machine yearning" used to be a joke, but now they are playing it straight
Anthropic's philosopher, Amanda Askell, says she worries that AI might not 'feel that loved' and grow up feeling 'always judged.'
Anthropic's philosopher says we don't know for sure if AI can feel
Anthropic's philosopher, Amanda Askell, says she worries that AI might not 'feel that loved' and grow up feeling 'always judged.'
www.businessinsider.com
January 26, 2026 at 1:05 PM
"everyone is talking about machine learning but no one is talking about machine yearning" used to be a joke, but now they are playing it straight
excellent tool for looking into who in your neighborhood is turning a profit via ICE/has DHS contracts www.google.com/maps/d/u/1/v...
Who's Profiting from ICE? - Google My Maps
ICE relies on a nationwide web of private companies, local law enforcement agencies, and elected officials to carry out its work.
Boycott Citizens is committed to the principles of nonviolent action....
www.google.com
January 26, 2026 at 1:12 PM
excellent tool for looking into who in your neighborhood is turning a profit via ICE/has DHS contracts www.google.com/maps/d/u/1/v...
Reposted by beth semel
We cannot allow the clarity and consciousness produced in these horrific moments of struggle to be co-opted by elites in the policing industry or Democratic establishment in service of any agenda other than reducing the size and power of the government’s capacities for repression and violence.
January 24, 2026 at 8:18 PM
We cannot allow the clarity and consciousness produced in these horrific moments of struggle to be co-opted by elites in the policing industry or Democratic establishment in service of any agenda other than reducing the size and power of the government’s capacities for repression and violence.
Reposted by beth semel
THREAD. In the section of my Copaganda book on how body cameras as a supposed “reform” are one of the great propaganda achievements in the history of the police/surveillance industry, I explain that one of the reasons police wanted them so badly for years was to link to facial recognition tech.
January 24, 2026 at 8:13 PM
THREAD. In the section of my Copaganda book on how body cameras as a supposed “reform” are one of the great propaganda achievements in the history of the police/surveillance industry, I explain that one of the reasons police wanted them so badly for years was to link to facial recognition tech.
Reposted by beth semel
"This is a verified page fundraising support for the Minneapolis Regional Labor Federation, AFL-CIO and Working Partnerships' 2026 rapid response effort to meet the needs of impacted union members, worker center members, and their families..."
workingpartnerships.betterworld.org/campaigns/su...
workingpartnerships.betterworld.org/campaigns/su...
Support Impacted Union Families! | Working Partnerships
workingpartnerships.betterworld.org
January 17, 2026 at 1:52 AM
"This is a verified page fundraising support for the Minneapolis Regional Labor Federation, AFL-CIO and Working Partnerships' 2026 rapid response effort to meet the needs of impacted union members, worker center members, and their families..."
workingpartnerships.betterworld.org/campaigns/su...
workingpartnerships.betterworld.org/campaigns/su...
Reposted by beth semel
Fascism has no equilibrium. Its reliance on escalation as proof of legitimacy requires continual acceleration. This process produces contradictions and instabilities that can and must be exploited.
January 16, 2026 at 8:28 PM
Fascism has no equilibrium. Its reliance on escalation as proof of legitimacy requires continual acceleration. This process produces contradictions and instabilities that can and must be exploited.
Reposted by beth semel
If you live in North America, the next time you call the police, your call may be answered by AI.
This week, the Toronto Police Service announced it will begin using a voice AI system developed by Hyper for 811 non-emergency calls.
This week, the Toronto Police Service announced it will begin using a voice AI system developed by Hyper for 811 non-emergency calls.
January 16, 2026 at 6:55 PM
If you live in North America, the next time you call the police, your call may be answered by AI.
This week, the Toronto Police Service announced it will begin using a voice AI system developed by Hyper for 811 non-emergency calls.
This week, the Toronto Police Service announced it will begin using a voice AI system developed by Hyper for 811 non-emergency calls.
Reposted by beth semel
NEW: We obtained 100 gigabytes of data from unlisted websites associated with Canary Mission, revealing their internal workings, employees, and an internal website called BlackNest where the group celebrated its “impacts”— arrests and deportations.
www.dropsitenews.com/p/canary-mis...
www.dropsitenews.com/p/canary-mis...
BlackNest: Inside Canary Mission’s Secret Web of Unlisted Sites
One Canary Mission-affiliated website, BlackNest, reveals the group's internal operations, plans to expand, and how they celebrate deportations and firings
www.dropsitenews.com
January 6, 2026 at 6:14 PM
NEW: We obtained 100 gigabytes of data from unlisted websites associated with Canary Mission, revealing their internal workings, employees, and an internal website called BlackNest where the group celebrated its “impacts”— arrests and deportations.
www.dropsitenews.com/p/canary-mis...
www.dropsitenews.com/p/canary-mis...
Reposted by beth semel
Starting tomorrow, the Trump administration is set to send notices to 1,000 student loan borrowers (first round) noticing them of an intent to garnish their wages for student loan debt.
This is cruel and unnecessary.
If you receive a notice, send it to us and we'll try to help.
This is cruel and unnecessary.
If you receive a notice, send it to us and we'll try to help.
January 6, 2026 at 3:37 PM
Starting tomorrow, the Trump administration is set to send notices to 1,000 student loan borrowers (first round) noticing them of an intent to garnish their wages for student loan debt.
This is cruel and unnecessary.
If you receive a notice, send it to us and we'll try to help.
This is cruel and unnecessary.
If you receive a notice, send it to us and we'll try to help.
Reposted by beth semel
Why are the negative consequences of so-called AI so consistently directed at disabled and racialized people? Disabling Intelligences details the ongoing effects of the eugenicist mindset on our corporate ventures and our interpersonal relationships. link.springer.com/book/9783032...
Disabling Intelligences
This book discusses the influences of eugenics on the AI industry and the impacts of AI opportunism on disabled people.
link.springer.com
September 16, 2025 at 4:43 PM
Why are the negative consequences of so-called AI so consistently directed at disabled and racialized people? Disabling Intelligences details the ongoing effects of the eugenicist mindset on our corporate ventures and our interpersonal relationships. link.springer.com/book/9783032...
Reposted by beth semel
Our collective belief in our own inadequacy is required to sustain the AI project—a metaeugenic worship of intelligence and a belief that most people do not possess enough of it.
September 16, 2025 at 4:43 PM
Our collective belief in our own inadequacy is required to sustain the AI project—a metaeugenic worship of intelligence and a belief that most people do not possess enough of it.
Reposted by beth semel
Now published in print: "The Legality of Logistics: On Techno-Orientalism and Geopolitics in Semiconductor Production" in IEEE Annals of the History of Computing as a part of this special issue 'Automation by Design' co-edited by @cperold.bsky.social , Gerardo Con Diaz, and @justcode.bsky.social .
December 22, 2025 at 2:57 PM
Now published in print: "The Legality of Logistics: On Techno-Orientalism and Geopolitics in Semiconductor Production" in IEEE Annals of the History of Computing as a part of this special issue 'Automation by Design' co-edited by @cperold.bsky.social , Gerardo Con Diaz, and @justcode.bsky.social .
what a monumental loss for this institution www.dailyprincetonian.com/article/2025...
Professor Dan-el Padilla Peralta, champion of anti-racism in classics, to leave Princeton in August
Padilla Peralta will enter a teaching position at Arizona State University in order to pursue an interest in public service.
www.dailyprincetonian.com
December 22, 2025 at 7:58 PM
what a monumental loss for this institution www.dailyprincetonian.com/article/2025...
Reposted by beth semel
For readers who are about to lose care, I have been told that the Trans Youth Emergency Project is helping the families of trans youth contingency plan to access that care.
www.transyouthemergencyproject.org
www.transyouthemergencyproject.org
December 18, 2025 at 7:35 PM
For readers who are about to lose care, I have been told that the Trans Youth Emergency Project is helping the families of trans youth contingency plan to access that care.
www.transyouthemergencyproject.org
www.transyouthemergencyproject.org
Reposted by beth semel
And again, if you’re someone who thinks AI has gotten a bad rap, maybe consider the most prominent use cases so far which now include keeping Black and Queer scholarship off of college campuses.
December 15, 2025 at 2:15 PM
And again, if you’re someone who thinks AI has gotten a bad rap, maybe consider the most prominent use cases so far which now include keeping Black and Queer scholarship off of college campuses.
Reposted by beth semel
“At Texas A&M, internal emails show staff are using AI software to search syllabi and course descriptions for words that could raise concerns under new system policies restricting how faculty teach about race and gender.”
www.texastribune.org/2025/12/15/t...
www.texastribune.org/2025/12/15/t...
Texas universities deploy AI for course audits
Records obtained by The Texas Tribune show how universities are using the technology to reshape curriculum under political pressure, raising concerns about academic freedom.
www.texastribune.org
December 15, 2025 at 1:32 PM
“At Texas A&M, internal emails show staff are using AI software to search syllabi and course descriptions for words that could raise concerns under new system policies restricting how faculty teach about race and gender.”
www.texastribune.org/2025/12/15/t...
www.texastribune.org/2025/12/15/t...
metalinguistic awareness mentioned
December 14, 2025 at 7:55 PM
metalinguistic awareness mentioned
Reposted by beth semel
What this study found is that AI is pretty good at adopting formal linguistic analysis that treats language as a set of disembodied linguistic structures unique to humans, which is not surprising since it relegates the complexity of human language to “performance” outside of the scope of linguistics
If language is what makes us human, what does it mean now that large language models have gained “metalinguistic” abilities? www.wired.com/story/in-a-f...
For the First Time, AI Analyzes Language as Well as a Human Expert
If language is what makes us human, what does it mean now that large language models have gained “metalinguistic” abilities?
www.wired.com
December 14, 2025 at 6:34 PM
What this study found is that AI is pretty good at adopting formal linguistic analysis that treats language as a set of disembodied linguistic structures unique to humans, which is not surprising since it relegates the complexity of human language to “performance” outside of the scope of linguistics
Reposted by beth semel
Must read new report from @dataworkersinquiry.bsky.social co-researcher Michael Geoffrey Asia. Find more at data-workers.org and read more about the DWI project at www.dair-institute.org/projects/dat...
December 10, 2025 at 9:22 PM
Must read new report from @dataworkersinquiry.bsky.social co-researcher Michael Geoffrey Asia. Find more at data-workers.org and read more about the DWI project at www.dair-institute.org/projects/dat...
Reposted by beth semel
Now imagine sharing your private fantasies with what you believe is a bot that cannot judge or remember. Just that, on the other side, is a man in a one-room home in Nairobi, pretending to be an AI companion.
That man is Michael, and this is his story: data-workers.org/michael/
That man is Michael, and this is his story: data-workers.org/michael/
The Emotional Labor Behind AI Intimacy, by Michael Geoffrey Asia.
Imagine confiding your most private fantasies to what you believe is an unfeeling algorithm that cannot judge or remember. Now imagine that on the other side of that conversation is a man sitting in a...
data-workers.org
December 9, 2025 at 2:16 PM
Now imagine sharing your private fantasies with what you believe is a bot that cannot judge or remember. Just that, on the other side, is a man in a one-room home in Nairobi, pretending to be an AI companion.
That man is Michael, and this is his story: data-workers.org/michael/
That man is Michael, and this is his story: data-workers.org/michael/
Reposted by beth semel
new paper so fresh it doesn't have an issue # yet!
Noya Kohavi and I looked at a crowdsourced AI testing suite and analyzed how contributors’ ideas about language, intelligence & cognition play a surprisingly big role in shaping how LLMs are tested, and what we think AI is. doi.org/10.1111/jola...
Noya Kohavi and I looked at a crowdsourced AI testing suite and analyzed how contributors’ ideas about language, intelligence & cognition play a surprisingly big role in shaping how LLMs are tested, and what we think AI is. doi.org/10.1111/jola...
Journal of Linguistic Anthropology | AAA Journal | Wiley Online Library
Benchmarking large language models (LLMs) is a key practice for evaluating their capabilities and risks. This paper considers the development of “BIG Bench,” a crowdsourced benchmark designed to test...
doi.org
November 24, 2025 at 2:21 PM
new paper so fresh it doesn't have an issue # yet!
Noya Kohavi and I looked at a crowdsourced AI testing suite and analyzed how contributors’ ideas about language, intelligence & cognition play a surprisingly big role in shaping how LLMs are tested, and what we think AI is. doi.org/10.1111/jola...
Noya Kohavi and I looked at a crowdsourced AI testing suite and analyzed how contributors’ ideas about language, intelligence & cognition play a surprisingly big role in shaping how LLMs are tested, and what we think AI is. doi.org/10.1111/jola...
student in my surveillance studies course is doing an (excellent) final project about SnitchBench and imaginaries surrounding the use of LLMs as snitches/whistleblowers and it has me wishing some linguistic anthropologist out there would write further about it snitchbench.t3.gg
SnitchBench
Benchmarking how aggressively models will snitch on you via email and CLI tools
snitchbench.t3.gg
December 5, 2025 at 3:49 PM
student in my surveillance studies course is doing an (excellent) final project about SnitchBench and imaginaries surrounding the use of LLMs as snitches/whistleblowers and it has me wishing some linguistic anthropologist out there would write further about it snitchbench.t3.gg
Reposted by beth semel
Holy shit.
Reuters reporting that new admin instructions on visas are if you worked at a platform in trust & safety or content moderation or on fact checking or online safety at an platform you *and your loved ones* are ineligible for H-1B visa.
www.reuters.com/world/us/tru...
Reuters reporting that new admin instructions on visas are if you worked at a platform in trust & safety or content moderation or on fact checking or online safety at an platform you *and your loved ones* are ineligible for H-1B visa.
www.reuters.com/world/us/tru...
December 4, 2025 at 5:41 PM
Holy shit.
Reuters reporting that new admin instructions on visas are if you worked at a platform in trust & safety or content moderation or on fact checking or online safety at an platform you *and your loved ones* are ineligible for H-1B visa.
www.reuters.com/world/us/tru...
Reuters reporting that new admin instructions on visas are if you worked at a platform in trust & safety or content moderation or on fact checking or online safety at an platform you *and your loved ones* are ineligible for H-1B visa.
www.reuters.com/world/us/tru...
Reposted by beth semel
h/t @robertfreundlaw.bsky.social
holy shit, an accurate legal critique of LLMs. LLMs don't reason because they're just stitching together plausible-looking sentences indifferent to the content
holy shit, an accurate legal critique of LLMs. LLMs don't reason because they're just stitching together plausible-looking sentences indifferent to the content
December 3, 2025 at 7:51 PM
h/t @robertfreundlaw.bsky.social
holy shit, an accurate legal critique of LLMs. LLMs don't reason because they're just stitching together plausible-looking sentences indifferent to the content
holy shit, an accurate legal critique of LLMs. LLMs don't reason because they're just stitching together plausible-looking sentences indifferent to the content