λTotoro
@lambdatotoro.bsky.social
Nonbinary CS PhD student, lambda enthusiast and science communicator. Equality. Freedom. Solidarity. ✊🏳️🌈🏴
Sometimes antisocial but always antifascist!
↙️↙️↙️
(any pronouns)
Sometimes antisocial but always antifascist!
↙️↙️↙️
(any pronouns)
Reposted by λTotoro
People stress me out with how much they misunderstand "alt text is contextual".
It doesn't mean you put the context of the image in the alt text. It doesn't mean you skimp on the alt text, either.
It means the alt text is going to change depending on where and why you post the image.
It doesn't mean you put the context of the image in the alt text. It doesn't mean you skimp on the alt text, either.
It means the alt text is going to change depending on where and why you post the image.
The main point of alt text is to say what the point of the image is in context. So that someone who can see it will get the same meaning as someone who is relying on a screen reader for alt text.
And because of that, where you're sharing the image will very probably change what you say it means.
And because of that, where you're sharing the image will very probably change what you say it means.
October 29, 2024 at 1:09 AM
People stress me out with how much they misunderstand "alt text is contextual".
It doesn't mean you put the context of the image in the alt text. It doesn't mean you skimp on the alt text, either.
It means the alt text is going to change depending on where and why you post the image.
It doesn't mean you put the context of the image in the alt text. It doesn't mean you skimp on the alt text, either.
It means the alt text is going to change depending on where and why you post the image.
Reposted by λTotoro
New insult to life itself just dropped
November 14, 2025 at 3:57 PM
New insult to life itself just dropped
Today's Mission: squeezing data from a stone. 🪨
November 11, 2025 at 10:56 AM
Today's Mission: squeezing data from a stone. 🪨
Reposted by λTotoro
Don't call the cops on someone in a mental health crisis unless you want them dead.
Three former San Antonio police officers were acquitted on Monday of all charges related to the fatal shooting of a woman who appeared to be having a mental health crisis when they were dispatched to her apartment in 2023. nyti.ms/4qRnvvT
3 San Antonio Officers Acquitted in Fatal Shooting of Woman in 2023
Melissa Perez, 46, swung a hammer in their direction and appeared to be in distress in her home when the officers arrived, officials said.
nyti.ms
November 10, 2025 at 10:38 PM
Don't call the cops on someone in a mental health crisis unless you want them dead.
I don't need it. I don't have a space for it. It will only ever gather dust. It's going to be expensive. I don't need it. Fuck Paramount, actually, I don't want to give them any money. It fits with nothing else I own. I don't need it. I don't want to fall for these nostalgia bait products. I don't n
![Article snippet: "in addition, for fans who purchase the Enterprise-D set over the first three days [...] the company will include a limited-edition 261-piece Shuttlepod Onizuka gift-with-purchase brickset which will include an exclusive Ensign Ro Laren minifigure."
Image shows said shuttlepod set and minifigure.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:gb3hsy46pcyityqfszfrfqhb/bafkreiafdrocccvnmekfihkhrvjwzaeraa2vfqzxea4fkjldtsh5634itm@jpeg)
November 7, 2025 at 4:42 PM
I don't need it. I don't have a space for it. It will only ever gather dust. It's going to be expensive. I don't need it. Fuck Paramount, actually, I don't want to give them any money. It fits with nothing else I own. I don't need it. I don't want to fall for these nostalgia bait products. I don't n
Reposted by λTotoro
Buckle up boys, tech bros are reinventing phrenology (again)
A new paper suggests a photo can tell a recruiter much about an applicant’s personality
Should facial analysis help determine whom companies hire?
A new paper suggests a photo can tell a recruiter much about an applicant’s personality
econ.st
November 6, 2025 at 11:29 PM
Buckle up boys, tech bros are reinventing phrenology (again)
Reposted by λTotoro
Oh, Dick Cheney died? Well-- *sees Bluesky mod around corner* What's really important is championing the open protocol.
November 4, 2025 at 6:06 PM
Oh, Dick Cheney died? Well-- *sees Bluesky mod around corner* What's really important is championing the open protocol.
Reposted by λTotoro
Post a bird announcing Dick Cheney is dead. 🦤
November 4, 2025 at 2:53 PM
Post a bird announcing Dick Cheney is dead. 🦤
Post a bird announcing Dick Cheney is dead. 🐦
November 4, 2025 at 12:24 PM
Post a bird announcing Dick Cheney is dead. 🐦
Computer Scientist's opinion:
No, for several reasons.
a) The parts where our civilisation is lacking are *social* rather than technological or scientific.
b) There can be no training data for things nobody has done before.
c) Science works best when things are true, which AI is incompatible with.
No, for several reasons.
a) The parts where our civilisation is lacking are *social* rather than technological or scientific.
b) There can be no training data for things nobody has done before.
c) Science works best when things are true, which AI is incompatible with.
Doesn’t it seem like an inevitable, and hugely beneficial asset to have AI involved in research? It seems like the next step towards making huge advancements as a civilization.
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 30, 2025 at 9:44 AM
Computer Scientist's opinion:
No, for several reasons.
a) The parts where our civilisation is lacking are *social* rather than technological or scientific.
b) There can be no training data for things nobody has done before.
c) Science works best when things are true, which AI is incompatible with.
No, for several reasons.
a) The parts where our civilisation is lacking are *social* rather than technological or scientific.
b) There can be no training data for things nobody has done before.
c) Science works best when things are true, which AI is incompatible with.
Reposted by λTotoro
My name is Ozymandias, King of Kings. Look on my works and let me know if you have any questions! 🤗
October 29, 2025 at 9:16 PM
My name is Ozymandias, King of Kings. Look on my works and let me know if you have any questions! 🤗
Reposted by λTotoro
Reposted by λTotoro
I'm generally a materialist about why most A.I. is so destructive, but I contain multitudes and so also believe that using A.I. is toxic for your one precious human soul, on a spiritual level.
October 27, 2025 at 7:33 PM
I'm generally a materialist about why most A.I. is so destructive, but I contain multitudes and so also believe that using A.I. is toxic for your one precious human soul, on a spiritual level.
Reposted by λTotoro
Remember this can also have some long-term consequences that are really serious. Certain diagnosis or medications can have implications on future employment as well as the ability to hold certain certifications and licenses. Getting your medical record corrected after this will be an nightmare
October 26, 2025 at 7:42 AM
Remember this can also have some long-term consequences that are really serious. Certain diagnosis or medications can have implications on future employment as well as the ability to hold certain certifications and licenses. Getting your medical record corrected after this will be an nightmare
Reposted by λTotoro
Yet again, we can't afford to let LLMs become a source of epistemic grounding for society.
Largest study of its kind shows AI assistants misrepresent news content 45% of the time – regardless of language or territory
An intensive international study was coordinated by the European Broadcasting Union (EBU) and led by the BBC
www.bbc.co.uk
October 24, 2025 at 5:21 AM
Yet again, we can't afford to let LLMs become a source of epistemic grounding for society.
I just had to give an important presentation for work and I precisely hit the mark of how long it was supposed to be. Neither too much content nor too little. And no major stumbles either.
I think I am ready to receive my medal now! 😌
I think I am ready to receive my medal now! 😌
October 22, 2025 at 1:44 PM
I just had to give an important presentation for work and I precisely hit the mark of how long it was supposed to be. Neither too much content nor too little. And no major stumbles either.
I think I am ready to receive my medal now! 😌
I think I am ready to receive my medal now! 😌
Reposted by λTotoro
We knew this would happen. Note how realistic the visuals and voices are.
This AI deepfake video of Catherine Connolly ‘withdrawing’ from presidential contest and that Friday’s poll is ‘cancelled’ is live and viral on Facebook tonight.
1000s watching, commenters believing it.
This AI deepfake video of Catherine Connolly ‘withdrawing’ from presidential contest and that Friday’s poll is ‘cancelled’ is live and viral on Facebook tonight.
1000s watching, commenters believing it.
October 21, 2025 at 9:58 PM
We knew this would happen. Note how realistic the visuals and voices are.
This AI deepfake video of Catherine Connolly ‘withdrawing’ from presidential contest and that Friday’s poll is ‘cancelled’ is live and viral on Facebook tonight.
1000s watching, commenters believing it.
This AI deepfake video of Catherine Connolly ‘withdrawing’ from presidential contest and that Friday’s poll is ‘cancelled’ is live and viral on Facebook tonight.
1000s watching, commenters believing it.
Reposted by λTotoro
okay hello. it's kate alt text clinic time. show me things you didn't write alt text for because you thought it was too difficult, or for whatever other reason, please. i will help
October 20, 2025 at 3:11 PM
okay hello. it's kate alt text clinic time. show me things you didn't write alt text for because you thought it was too difficult, or for whatever other reason, please. i will help
Reposted by λTotoro
Once again: The premise of AI is that life would be great if you didn't have to live it.
A new iPhone app, Endless Summer, uses AI to create photorealistic vacation photos starring you — no travel required
Too burned out to travel? This new app fakes your summer vacation photos for you | TechCrunch
A new iPhone app, Endless Summer, uses AI to create photorealistic vacation photos starring you — no travel required
techcrunch.com
October 21, 2025 at 9:28 AM
Once again: The premise of AI is that life would be great if you didn't have to live it.
Reposted by λTotoro
everything is capybara
October 20, 2025 at 8:49 PM
everything is capybara
Reposted by λTotoro
My mistake, you are correct! I generated the file, but I did not give you a download link — let me go get it for you [pretends to walk down stairs behind couch]
This person is paying OpenAI two hundred dollars a month to have a chatbot gaslight them about making an excel spreadsheet
October 20, 2025 at 2:22 AM
My mistake, you are correct! I generated the file, but I did not give you a download link — let me go get it for you [pretends to walk down stairs behind couch]
Reposted by λTotoro
Nothin else for it
October 20, 2025 at 4:02 PM
Nothin else for it
Reposted by λTotoro
Incels were the easy mode.
Hard mode is incels who've spent 10,000 hours learning what women are like by talking to a billionaire sex pest's gacha machine disguised as a fawning sycophant with no agency with a suspiciously child-like face.
Hard mode is incels who've spent 10,000 hours learning what women are like by talking to a billionaire sex pest's gacha machine disguised as a fawning sycophant with no agency with a suspiciously child-like face.
July 17, 2025 at 6:24 AM
Incels were the easy mode.
Hard mode is incels who've spent 10,000 hours learning what women are like by talking to a billionaire sex pest's gacha machine disguised as a fawning sycophant with no agency with a suspiciously child-like face.
Hard mode is incels who've spent 10,000 hours learning what women are like by talking to a billionaire sex pest's gacha machine disguised as a fawning sycophant with no agency with a suspiciously child-like face.
Reposted by λTotoro
When I can’t find the answer to something, if I figure it out, it’s my duty to tell others.
This includes calling back customer support to tell them an answer I found, so they can help future callers.
This includes calling back customer support to tell them an answer I found, so they can help future callers.
Do you have any extremely niche, but serious, ethical stances?
October 19, 2025 at 6:50 PM
When I can’t find the answer to something, if I figure it out, it’s my duty to tell others.
This includes calling back customer support to tell them an answer I found, so they can help future callers.
This includes calling back customer support to tell them an answer I found, so they can help future callers.