Eddie Antonio Santos
@eddieantonio.ca
I love teaching people how to code. PhD researcher. (he/him)
Reposted by Eddie Antonio Santos
🚨A.I. INFILTRATION ALERT🚨
If you use Gmail (which is a LOT of you), the fuckin' company just installed Gemini into all of your "smart features," thus scanning all of your emails
You have to turn it off, then manually re-enable your smart inbox
It's a pain in the ass, but it's worth both steps:
If you use Gmail (which is a LOT of you), the fuckin' company just installed Gemini into all of your "smart features," thus scanning all of your emails
You have to turn it off, then manually re-enable your smart inbox
It's a pain in the ass, but it's worth both steps:
November 21, 2025 at 9:20 AM
🚨A.I. INFILTRATION ALERT🚨
If you use Gmail (which is a LOT of you), the fuckin' company just installed Gemini into all of your "smart features," thus scanning all of your emails
You have to turn it off, then manually re-enable your smart inbox
It's a pain in the ass, but it's worth both steps:
If you use Gmail (which is a LOT of you), the fuckin' company just installed Gemini into all of your "smart features," thus scanning all of your emails
You have to turn it off, then manually re-enable your smart inbox
It's a pain in the ass, but it's worth both steps:
Reposted by Eddie Antonio Santos
LLMs destroy context. You used to be able to tell a crank website from a good source partly by its style.
But now we have a tool that puts careful research, incoherent garbage, and intentional shitposting through a kind of grammatical instagram filter, presenting them all with the same credibility
But now we have a tool that puts careful research, incoherent garbage, and intentional shitposting through a kind of grammatical instagram filter, presenting them all with the same credibility
The bigger change is presentation, historically, these emails were visually chaotic: sudden colour changes, ALL CAPS, screenshots pasted at random. Classic signs of disordered thinking. Now they’re neat, structured, and grammatically consistent. The delusion hasn't changed, but the formatting has.
November 19, 2025 at 4:12 PM
LLMs destroy context. You used to be able to tell a crank website from a good source partly by its style.
But now we have a tool that puts careful research, incoherent garbage, and intentional shitposting through a kind of grammatical instagram filter, presenting them all with the same credibility
But now we have a tool that puts careful research, incoherent garbage, and intentional shitposting through a kind of grammatical instagram filter, presenting them all with the same credibility
Reposted by Eddie Antonio Santos
engineers are basically professional understanders of complex systems. When someone I respect sends a PR I trust that it reflects a reasonably complete model of the system in their head. AI doesn't have it, it has an educated guess with varying levels of plausibility.
November 15, 2025 at 12:41 AM
engineers are basically professional understanders of complex systems. When someone I respect sends a PR I trust that it reflects a reasonably complete model of the system in their head. AI doesn't have it, it has an educated guess with varying levels of plausibility.
Reposted by Eddie Antonio Santos
the amount of people who will say something like "they went to tech so you know they're smart"
YALL
math science engineering is a certain kind of skill that some ppl study but lolll plz do not defer to stem folks on all the things, just on stem things
YALL
math science engineering is a certain kind of skill that some ppl study but lolll plz do not defer to stem folks on all the things, just on stem things
November 10, 2025 at 4:01 PM
the amount of people who will say something like "they went to tech so you know they're smart"
YALL
math science engineering is a certain kind of skill that some ppl study but lolll plz do not defer to stem folks on all the things, just on stem things
YALL
math science engineering is a certain kind of skill that some ppl study but lolll plz do not defer to stem folks on all the things, just on stem things
Reposted by Eddie Antonio Santos
Who made this and why are you choosing nerd violence? ✍️ I mean, it is true, but still.
November 9, 2025 at 6:41 PM
Who made this and why are you choosing nerd violence? ✍️ I mean, it is true, but still.
Reposted by Eddie Antonio Santos
Turn off Gemini in Google!
It's been turned on by default.
Go to your Drive, click the gear icon, click settings, go to "manage apps" and uncheck that nasty "use by default" box.
www.zdnet.com/article/how-...
It's been turned on by default.
Go to your Drive, click the gear icon, click settings, go to "manage apps" and uncheck that nasty "use by default" box.
www.zdnet.com/article/how-...
How to turn off Gemini in your Gmail, Docs, Photos, and more - it's easy to opt out
It's a little hidden, but there is a way to remove Gemini from your favorite Google services.
www.zdnet.com
October 27, 2025 at 7:10 PM
Turn off Gemini in Google!
It's been turned on by default.
Go to your Drive, click the gear icon, click settings, go to "manage apps" and uncheck that nasty "use by default" box.
www.zdnet.com/article/how-...
It's been turned on by default.
Go to your Drive, click the gear icon, click settings, go to "manage apps" and uncheck that nasty "use by default" box.
www.zdnet.com/article/how-...
Reposted by Eddie Antonio Santos
Bréaga a chreideann ríomhchláraitheoirí faoi teangacha, le samplaí. Ó @lexiconista.bsky.social
www.lexiconista.com/falsehoods-a...
www.lexiconista.com/falsehoods-a...
Falsehoods programmers believe about languages
This is what we have to put up with in the software localisation industry.
www.lexiconista.com
October 13, 2025 at 5:15 PM
Bréaga a chreideann ríomhchláraitheoirí faoi teangacha, le samplaí. Ó @lexiconista.bsky.social
www.lexiconista.com/falsehoods-a...
www.lexiconista.com/falsehoods-a...
Reposted by Eddie Antonio Santos
“Well-meaning” people are using machine translation to write Wikipedia articles in languages that they don’t speak themselves, accelerating the degeneration in quality of the web corpus for several languages with relatively few native speakers.
www.technologyreview.com/2025/09/25/1...
www.technologyreview.com/2025/09/25/1...
How AI and Wikipedia have sent vulnerable languages into a doom spiral
Machine translators have made it easier than ever to create error-plagued Wikipedia articles in obscure languages. What happens when AI models get trained on junk pages?
www.technologyreview.com
October 16, 2025 at 3:14 AM
“Well-meaning” people are using machine translation to write Wikipedia articles in languages that they don’t speak themselves, accelerating the degeneration in quality of the web corpus for several languages with relatively few native speakers.
www.technologyreview.com/2025/09/25/1...
www.technologyreview.com/2025/09/25/1...
Reposted by Eddie Antonio Santos
I've written about the life and just some of the achievements of my former classmate Andy West aka 魏安, Ando, and BabelStone @babelstone.co.uk
www.theguardian.com/technology/2...
www.theguardian.com/technology/2...
Andy West obituary
Other lives: Expert in encoding scripts for languages
www.theguardian.com
October 9, 2025 at 7:03 AM
I've written about the life and just some of the achievements of my former classmate Andy West aka 魏安, Ando, and BabelStone @babelstone.co.uk
www.theguardian.com/technology/2...
www.theguardian.com/technology/2...
Reposted by Eddie Antonio Santos
Meta-analysis of 262 articles showing robust links between the vitality of Indigenous languages and health or wellness.
The mental health benefits are particularly notable. Language programs help Indigenous communities thrive.
www.sciencedirect.com/science/arti...
The mental health benefits are particularly notable. Language programs help Indigenous communities thrive.
www.sciencedirect.com/science/arti...
Language improves health and wellbeing in Indigenous communities: A scoping review
Indigenous languages in Australia, Canada, New Zealand and the United States are endangered due to colonial policies which promote English language do…
www.sciencedirect.com
September 24, 2025 at 2:53 AM
Meta-analysis of 262 articles showing robust links between the vitality of Indigenous languages and health or wellness.
The mental health benefits are particularly notable. Language programs help Indigenous communities thrive.
www.sciencedirect.com/science/arti...
The mental health benefits are particularly notable. Language programs help Indigenous communities thrive.
www.sciencedirect.com/science/arti...
Reposted by Eddie Antonio Santos
If you’re on academia dot edu, let me suggest that you strongly consider deleting your account.
September 17, 2025 at 7:09 PM
If you’re on academia dot edu, let me suggest that you strongly consider deleting your account.
Reposted by Eddie Antonio Santos
Happy Unicode 17.0 Day!
Includes seven new emoji that are rolling out on your phones and computers in the coming year:
treasure chest, trombone, avalanche, big foot, bulging face, fight cloud, orca
@jenniferdaniel.bsky.social has the lowdown:
jenniferdaniel.substack.com/p/tomorrows-...
Includes seven new emoji that are rolling out on your phones and computers in the coming year:
treasure chest, trombone, avalanche, big foot, bulging face, fight cloud, orca
@jenniferdaniel.bsky.social has the lowdown:
jenniferdaniel.substack.com/p/tomorrows-...
September 10, 2025 at 3:22 AM
Happy Unicode 17.0 Day!
Includes seven new emoji that are rolling out on your phones and computers in the coming year:
treasure chest, trombone, avalanche, big foot, bulging face, fight cloud, orca
@jenniferdaniel.bsky.social has the lowdown:
jenniferdaniel.substack.com/p/tomorrows-...
Includes seven new emoji that are rolling out on your phones and computers in the coming year:
treasure chest, trombone, avalanche, big foot, bulging face, fight cloud, orca
@jenniferdaniel.bsky.social has the lowdown:
jenniferdaniel.substack.com/p/tomorrows-...
Reposted by Eddie Antonio Santos
Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia:
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
September 6, 2025 at 8:13 AM
Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia:
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
doi.org/10.5281/zeno...
We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n
Introduced my non-CS majors to floating point numbers yesterday by showing off .1 + .2, then showing them 0.30000000000000004.com and I think I lost a lot of them. But at the same time, I hate telling that lie that "floats are just numbers with a decimal". Does anybody have experience with this?
Floating Point Math
0.30000000000000004.com
September 5, 2025 at 1:33 PM
Introduced my non-CS majors to floating point numbers yesterday by showing off .1 + .2, then showing them 0.30000000000000004.com and I think I lost a lot of them. But at the same time, I hate telling that lie that "floats are just numbers with a decimal". Does anybody have experience with this?
Reposted by Eddie Antonio Santos
Having gotten through the first two weeks of classes, it's hard to convey just how radical my pedagogy has become, especially post-pandemic. Sure, AI and learning outcomes and other intrusions, but more just rethinking some of the cop shit that goes on in classrooms (including mine, in the past).
September 5, 2025 at 1:25 PM
Having gotten through the first two weeks of classes, it's hard to convey just how radical my pedagogy has become, especially post-pandemic. Sure, AI and learning outcomes and other intrusions, but more just rethinking some of the cop shit that goes on in classrooms (including mine, in the past).
Reposted by Eddie Antonio Santos
*cracks knuckles*
okay so you know how the media has a problem with taking snippets of scientific studies out of context and writing articles which mislead the public?
now imagine a computer program is doing that when you ask it questions about scientific literature but also it makes stuff up.
okay so you know how the media has a problem with taking snippets of scientific studies out of context and writing articles which mislead the public?
now imagine a computer program is doing that when you ask it questions about scientific literature but also it makes stuff up.
September 2, 2025 at 8:41 PM
*cracks knuckles*
okay so you know how the media has a problem with taking snippets of scientific studies out of context and writing articles which mislead the public?
now imagine a computer program is doing that when you ask it questions about scientific literature but also it makes stuff up.
okay so you know how the media has a problem with taking snippets of scientific studies out of context and writing articles which mislead the public?
now imagine a computer program is doing that when you ask it questions about scientific literature but also it makes stuff up.
Reposted by Eddie Antonio Santos
When reading AI reasoning text (aka CoT), we (humans) form a narrative about the underlying computation process, which we take as a transparent explanation of model behavior. But what if our narratives are wrong? We measure that and find it usually is.
Now on arXiv: arxiv.org/abs/2508.16599
Now on arXiv: arxiv.org/abs/2508.16599
Humans Perceive Wrong Narratives from AI Reasoning Texts
A new generation of AI models generates step-by-step reasoning text before producing an answer. This text appears to offer a human-readable window into their computation process, and is increasingly r...
arxiv.org
August 27, 2025 at 9:30 PM
When reading AI reasoning text (aka CoT), we (humans) form a narrative about the underlying computation process, which we take as a transparent explanation of model behavior. But what if our narratives are wrong? We measure that and find it usually is.
Now on arXiv: arxiv.org/abs/2508.16599
Now on arXiv: arxiv.org/abs/2508.16599
Reposted by Eddie Antonio Santos
Let's be clear: all the AI in edtech stuff (Canvas, Blackboard, Grammarly, Duolingo, OpenAI, the rest) is all about the "lock-in" - locking in students as subscribers or institutions via long-term license agreements for value generation. Is there good pedagogic evidence for it? No.
August 22, 2025 at 8:46 PM
Let's be clear: all the AI in edtech stuff (Canvas, Blackboard, Grammarly, Duolingo, OpenAI, the rest) is all about the "lock-in" - locking in students as subscribers or institutions via long-term license agreements for value generation. Is there good pedagogic evidence for it? No.
Reposted by Eddie Antonio Santos
Did you know that ELP offers a FREE Language Revitalization Mentors program?
Anyone who’s working to revitalize their language, wanting to get started, or curious about language revitalization, can talk one-on-one with experienced language revitalizers and scholars.
endangeredlanguages.com/mentors
Anyone who’s working to revitalize their language, wanting to get started, or curious about language revitalization, can talk one-on-one with experienced language revitalizers and scholars.
endangeredlanguages.com/mentors
May 9, 2025 at 4:08 PM
Did you know that ELP offers a FREE Language Revitalization Mentors program?
Anyone who’s working to revitalize their language, wanting to get started, or curious about language revitalization, can talk one-on-one with experienced language revitalizers and scholars.
endangeredlanguages.com/mentors
Anyone who’s working to revitalize their language, wanting to get started, or curious about language revitalization, can talk one-on-one with experienced language revitalizers and scholars.
endangeredlanguages.com/mentors
Reposted by Eddie Antonio Santos
"Age verification" laws are actually "upload your ID or get your face scanned to access every website, ending anonymity and associating your identity with everything you do online" laws and if more people understood that they would not be down for this authoritarian nonsense
April 2, 2025 at 12:57 PM
"Age verification" laws are actually "upload your ID or get your face scanned to access every website, ending anonymity and associating your identity with everything you do online" laws and if more people understood that they would not be down for this authoritarian nonsense
Reposted by Eddie Antonio Santos
When you struggle to find the equivalent of a word or phrase in another language, do you sometimes find yourself opening Wikipedia, searching for it then navigating to the "languages" section to get its title in the other language?
Well, I do, and I got tired of it, hence:
#linguistics #xl8 #xl9n
Well, I do, and I got tired of it, hence:
#linguistics #xl8 #xl9n
WikiXl8
Look up Wikipedia pages across languages
corpustools.prendrelangue.fr
August 20, 2025 at 9:09 AM
When you struggle to find the equivalent of a word or phrase in another language, do you sometimes find yourself opening Wikipedia, searching for it then navigating to the "languages" section to get its title in the other language?
Well, I do, and I got tired of it, hence:
#linguistics #xl8 #xl9n
Well, I do, and I got tired of it, hence:
#linguistics #xl8 #xl9n
Reposted by Eddie Antonio Santos
It really does feel like a lot of formerly reasonable people have been infected by some sort of terrifying brain parasite when it comes to unthinking and total acceptance of using AI tools
Absolute worst part, imo, is the pressure from school administrators to "welcome our new overlords" by adopting AI. It's also happening in elite universities, to my utter bafflement & horror. Anyone who calls bullshit on AI is treated as if only ignorance could explain their perspective.
August 8, 2025 at 11:38 AM
It really does feel like a lot of formerly reasonable people have been infected by some sort of terrifying brain parasite when it comes to unthinking and total acceptance of using AI tools
Reposted by Eddie Antonio Santos
Lit review query for #NLP folks: What are your favorite works reflecting on evaluation, especially of (UD) parsing and (FST or other) morphological analyzers?
August 6, 2025 at 8:41 PM
Lit review query for #NLP folks: What are your favorite works reflecting on evaluation, especially of (UD) parsing and (FST or other) morphological analyzers?
Reposted by Eddie Antonio Santos
Pretty interesting study! This result is going to get misinterpreted in a bunch of different ways over the next few days, so I thought I'd share a few thoughts with you now to contextualise some of the stuff in the paper.
We ran a randomized controlled trial to see how much AI coding tools speed up experienced open-source developers.
The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't.
The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't.
July 11, 2025 at 10:49 AM
Pretty interesting study! This result is going to get misinterpreted in a bunch of different ways over the next few days, so I thought I'd share a few thoughts with you now to contextualise some of the stuff in the paper.
Reposted by Eddie Antonio Santos
Are you able to use your language online in all the ways you want to? Do you have keyboards you can easily type your language in?
If the answer is "no", ELP Governance Council member Dr. Craig Cornelius can help for FREE! Contact the Keyboard Creation Project:
If the answer is "no", ELP Governance Council member Dr. Craig Cornelius can help for FREE! Contact the Keyboard Creation Project:
Keyboard Creation Project - Translation Commons
Keyboard Creation Project Translation Commons has assembled a group of volunteers from industry and academia to assist Indigenous communities to digitize their language. One of the first necessary ste...
translationcommons.org
June 30, 2025 at 7:34 PM
Are you able to use your language online in all the ways you want to? Do you have keyboards you can easily type your language in?
If the answer is "no", ELP Governance Council member Dr. Craig Cornelius can help for FREE! Contact the Keyboard Creation Project:
If the answer is "no", ELP Governance Council member Dr. Craig Cornelius can help for FREE! Contact the Keyboard Creation Project: