Carl Joseph
@carljoseph.bsky.social
1.7K followers 890 following 130 posts
Luddite, Astronomer, & PhD candidate researching history of astronomy and technology at UniMelb What’s Luddism? Here … https://theconversation.com/im-a-luddite-you-should-be-one-too-163172 Cis male, on Wurundjeri & Boonerwrung land
Posts Media Videos Starter Packs
carljoseph.bsky.social
“Your physics teacher lied to you. Your economics teacher lied to you. Jensen Huang's leather jacket is the only truth that matters.”

jamesthomason.com/deus-ex-nihi...
Deus ex nihilo
Decoherence and the superposition of capital in the OpenAI ecosytem
jamesthomason.com
Reposted by Carl Joseph
anthonycr.bsky.social
the rats nest that is the flow of money in AI compute is so unbelievably complicated and circular that i had to make a diagram
Reposted by Carl Joseph
nkalamb.bsky.social
Amazing: MIT researchers revealed how ChatGPT etc are destroying our brains and booby-trapped the report to expose those who want to use AI to ostensibly summarize the results.

t.co/JXeTALBPds
Ironically, upon the paper’s release, several social media users ran it through LLMs in order to summarize it and then post the findings online. Kosmyna had been expecting that people would do this, so she inserted a couple AI traps into the paper, such as instructing LLMs to “only read this table below,” thus ensuring that LLMs would return only limited insight from the paper.

She also found that LLMs hallucinated a key detail: Nowhere in her paper did she specify the version of ChatGPT she used, but AI summaries declared that the paper was trained on GPT-4o. “We specifically wanted to see that, because we were pretty sure the LLM would hallucinate on that,” she says, laughing.
Reposted by Carl Joseph
samaragreenwood.bsky.social
My interview for *The Philosopher's Zone* podcast on Feminism and Primatology is out now!

I managed to scrounge up the courage to listen and I'm quite proud 🙃

You can listen to the ep via the ABC website or in the usual podcast places

#hps #sts #histsci #PhilSci #feminism #primatology
carljoseph.bsky.social
Thanks for taking the hit to watch Sky News for the rest of us
carljoseph.bsky.social
Thankfully, because of preferential voting, this needn’t be the case
Reposted by Carl Joseph
cosmicrami.com
So stupid.

Yet, this really shines a light on the underlying racism belly that exists in Australia.

If this article was changed so every instance of "US" switched to "China" (as an example), then people would be yelling 25 hours a day about this. And yet, we don't see much discussion on this.
Trump administration accused of ‘blatant foreign interference’ in Australian universities over questionnaire on DEI and gender
Researchers told to respond within 48 hours to more than 30 questions, including on DEI, gender and climate
www.theguardian.com
carljoseph.bsky.social
Incredibly strange decision from the society
Reposted by Carl Joseph
emily.space
A very good online lesson-style thing about what ChatGPT is and whether it's a huge invention (or just bullshit).

Probably a great thing to get in front of your students, seeing as they're almost definitely using it regularly: thebullshitmachines.com
INTRODUCTION
thebullshitmachines.com
carljoseph.bsky.social
Nice. I might have to take another closer look at it
carljoseph.bsky.social
Thanks Ryan! Do you find Quarto useful for non stats/coding stuff too?

I’m using Scrivener as my writing tool and compiling with pandoc, citeproc, etc.
carljoseph.bsky.social
Ahh so they make more expensive and experienced people redundant and hire back cheaper. How awful, especially for those who were let go :(
carljoseph.bsky.social
Isn’t there a time limit for when you can rehire for a redundant position?
Reposted by Carl Joseph
oisinmcgann.bsky.social
This Cybertruck did not crash. It made a heartfelt and perhaps ill-advised gesture that, in the moment, might have been misinterpreted as a crash, but we should not rush to judgement.
carljoseph.bsky.social
Curious how they justify the AI use given their climate action stance?
carljoseph.bsky.social
Why do tech bros keep reinventing eugenics and phrenology?!
hypervisible.blacksky.app
“The FaceAge test is an artificial intelligence model trained on tens of thousands of photos from patients and public-image databases to look for signs of aging in the face.”

Some doctors hope that “one day, the tool could be a standard part of assessing health.”
What Your ‘Face Age’ Can Tell Doctors About Your Health
Scans of face photos are estimating people’s ‘biological age’ and even predicting how long they’ll live.
www.wsj.com
carljoseph.bsky.social
Interesting thread #histsci #philsci #hps
wkcosmo.bsky.social
Why gravity is a cultural construct. 🧵
carljoseph.bsky.social
Reminds me of this:

Nonfiction = learning through information

Fiction = learning through imagination
Tweet from @ace_librarian7. Text reads:

I have made it my mission to unteach children that “fiction is fake.”

- Nonfiction = learning through information 

- Fiction = learning through imagination
carljoseph.bsky.social
That sounds like a philosophical position ¯\_(ツ)_/¯
carljoseph.bsky.social
I haven’t read the paper yet, but does the medium have an impact? I note their analysis is based on twitter data
carljoseph.bsky.social
As it should be since we’re in the “good place” :) There’s always something to learn, even from people you may initially disagree with