New computational analyses on algorithmic bias from @timothyjgraham.bsky.social suggest that the X algorithm substantially boosted Republican accounts right after Elon Musk's endorsement of Donald Trump.
Full report eprints.qut.edu.au/253211/1/A_c...
Full report eprints.qut.edu.au/253211/1/A_c...
November 16, 2024 at 4:19 PM
New computational analyses on algorithmic bias from @timothyjgraham.bsky.social suggest that the X algorithm substantially boosted Republican accounts right after Elon Musk's endorsement of Donald Trump.
Full report eprints.qut.edu.au/253211/1/A_c...
Full report eprints.qut.edu.au/253211/1/A_c...
Nothing new really, Musk's been shamelessly manipulating X since he endorsed Trump post assassination attempt. At this point it's basically a GOP echo chamber.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
May 14, 2025 at 8:27 PM
Nothing new really, Musk's been shamelessly manipulating X since he endorsed Trump post assassination attempt. At this point it's basically a GOP echo chamber.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
qui lo hai tutto eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au
November 16, 2024 at 8:10 PM
qui lo hai tutto eprints.qut.edu.au/253211/1/A_c...
Musk ayudando en X a Trump, con datos: eprints.qut.edu.au/253211/1/A_c... Via @anneapplebaum.bsky.social
November 17, 2024 at 10:22 AM
Musk ayudando en X a Trump, con datos: eprints.qut.edu.au/253211/1/A_c... Via @anneapplebaum.bsky.social
TA : "on m'accuse de faire mon autopromo, mais moi j'ai pas besoin de vous pour faire ma promo, j'ai 44 reels qui sortent en juillet"
A_c : "c'est une machine de guerre la meuf hein faut le dire !"
A_c : "c'est une machine de guerre la meuf hein faut le dire !"
July 3, 2025 at 12:18 PM
TA : "on m'accuse de faire mon autopromo, mais moi j'ai pas besoin de vous pour faire ma promo, j'ai 44 reels qui sortent en juillet"
A_c : "c'est une machine de guerre la meuf hein faut le dire !"
A_c : "c'est une machine de guerre la meuf hein faut le dire !"
Muy buen video, buen diagnóstico, por fin alguien que entendió realmente el ensayo sobre heteropesimismo y buen análisis de cómo funcionan las redes sociales
youtu.be/A_C-G9M_0NI?...
youtu.be/A_C-G9M_0NI?...
TikTok Feminism is Not Your Friend
YouTube video by Broey Deschanel
youtu.be
March 1, 2026 at 7:09 PM
Muy buen video, buen diagnóstico, por fin alguien que entendió realmente el ensayo sobre heteropesimismo y buen análisis de cómo funcionan las redes sociales
youtu.be/A_C-G9M_0NI?...
youtu.be/A_C-G9M_0NI?...
Achei bem bom (fundamental minha opinião sobre feminismo e tik tok, dois assuntos que eu domino)
youtu.be/A_C-G9M_0NI?...
youtu.be/A_C-G9M_0NI?...
TikTok Feminism is Not Your Friend
YouTube video by Broey Deschanel
youtu.be
January 28, 2026 at 3:02 AM
Achei bem bom (fundamental minha opinião sobre feminismo e tik tok, dois assuntos que eu domino)
youtu.be/A_C-G9M_0NI?...
youtu.be/A_C-G9M_0NI?...
Es gibt genug Studien. Zum Beispiel dass auf X von Musks Trump endorsment an gezielt entsprechende Inhalte promotet wurden.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
December 13, 2024 at 12:15 PM
Es gibt genug Studien. Zum Beispiel dass auf X von Musks Trump endorsment an gezielt entsprechende Inhalte promotet wurden.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
i'm at a loss of words after reading a paper about _reformatting code_ using an ML model that has a measured statistical quantity A_c which says how often the reformatted code behaves the same as the original
the "ideal" (their choice of words) case is 64.2%
the "ideal" (their choice of words) case is 64.2%
March 23, 2026 at 11:03 PM
i'm at a loss of words after reading a paper about _reformatting code_ using an ML model that has a measured statistical quantity A_c which says how often the reformatted code behaves the same as the original
the "ideal" (their choice of words) case is 64.2%
the "ideal" (their choice of words) case is 64.2%
you could probably also say this video is about the female loneliness epidemic
www.youtube.com/watch?v=A_C-...
www.youtube.com/watch?v=A_C-...
TikTok Feminism is Not Your Friend
YouTube video by Broey Deschanel
www.youtube.com
January 26, 2026 at 9:59 PM
you could probably also say this video is about the female loneliness epidemic
www.youtube.com/watch?v=A_C-...
www.youtube.com/watch?v=A_C-...
I watched this youtube video today youtu.be/A_C-G9M_0NI?... and thought about how isolated we all are. How we've been made that way by our anti-human tech oligarchs sucking the life out of us. How it's greatest desire is to have us all isolated and angry and I made this painting in response.
TikTok Feminism is Not Your Friend
YouTube video by Broey Deschanel
youtu.be
February 24, 2026 at 4:01 AM
I watched this youtube video today youtu.be/A_C-G9M_0NI?... and thought about how isolated we all are. How we've been made that way by our anti-human tech oligarchs sucking the life out of us. How it's greatest desire is to have us all isolated and angry and I made this painting in response.
Graham, Timothy & Andrejevic, Mark
(2024)
A computational analysis of potential algorithmic bias on platform X during
the 2024 US election.
eprints.qut.edu.au/253211/1/A_c...
(2024)
A computational analysis of potential algorithmic bias on platform X during
the 2024 US election.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au
November 18, 2024 at 4:13 PM
Graham, Timothy & Andrejevic, Mark
(2024)
A computational analysis of potential algorithmic bias on platform X during
the 2024 US election.
eprints.qut.edu.au/253211/1/A_c...
(2024)
A computational analysis of potential algorithmic bias on platform X during
the 2024 US election.
eprints.qut.edu.au/253211/1/A_c...
How I learned about MUMPS:
thedailywtf.com/articles/A_C...
thedailywtf.com/articles/A_C...
February 12, 2025 at 11:25 PM
How I learned about MUMPS:
thedailywtf.com/articles/A_C...
thedailywtf.com/articles/A_C...
This analysis pinpoints the exact day - July 12th 2024 - when the X algorithm was tweaked to artificially boost both Elon Musk's account and the accounts of Republican politicians, in the lead-up to the US election
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
November 14, 2024 at 9:39 AM
This analysis pinpoints the exact day - July 12th 2024 - when the X algorithm was tweaked to artificially boost both Elon Musk's account and the accounts of Republican politicians, in the lead-up to the US election
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
Heavy bot usage : eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au
November 22, 2024 at 2:10 AM
Heavy bot usage : eprints.qut.edu.au/253211/1/A_c...
watched this video last night which helped explain how this current bizarre and purely-aesthetic strain of pop "feminism" came to be www.youtube.com/watch?v=A_C-...
TikTok Feminism is Not Your Friend
YouTube video by Broey Deschanel
www.youtube.com
January 29, 2026 at 4:05 PM
watched this video last night which helped explain how this current bizarre and purely-aesthetic strain of pop "feminism" came to be www.youtube.com/watch?v=A_C-...
TikTok Feminism is Not Your Friend
YouTube video by Broey Deschanel
youtu.be
April 17, 2026 at 12:07 AM
Nach der ungenauen Vance-NATO-Empörung erwartbar, dass deutsche Medien sich diese australische Studie zur möglichen Algorithmus-Veränderung durch Musk schnappen, ohne auf ihre löchrige Methodik oder die (bspw. auf X florierende) fundierte Kritik daran einzugehen.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au
November 19, 2024 at 7:41 AM
Nach der ungenauen Vance-NATO-Empörung erwartbar, dass deutsche Medien sich diese australische Studie zur möglichen Algorithmus-Veränderung durch Musk schnappen, ohne auf ihre löchrige Methodik oder die (bspw. auf X florierende) fundierte Kritik daran einzugehen.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
Some data points I would be looking for if I wanted to monitor whether X was boosting the Musk-AfD debate tonight, in @politico.eu with @pieterhaeck.bsky.social pro.politico.eu/news/191916. See also this study by @timothyjgraham.bsky.social eprints.qut.edu.au/253211/1/A_c...
January 9, 2025 at 8:36 AM
Some data points I would be looking for if I wanted to monitor whether X was boosting the Musk-AfD debate tonight, in @politico.eu with @pieterhaeck.bsky.social pro.politico.eu/news/191916. See also this study by @timothyjgraham.bsky.social eprints.qut.edu.au/253211/1/A_c...
4. Por último: Twitter. Estudios preliminares sugieren que los algoritmos se manipularon para favorecer al partido Republicano. Surprise, surprise. ¿Qué impacto tuvo? ¿Qué implicaciones tiene de cara a la comunicación política e institucional?
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
November 15, 2024 at 6:48 PM
4. Por último: Twitter. Estudios preliminares sugieren que los algoritmos se manipularon para favorecer al partido Republicano. Surprise, surprise. ¿Qué impacto tuvo? ¿Qué implicaciones tiene de cara a la comunicación política e institucional?
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
Izgleda da je Musk sredinom godine prilagodio X algoritam kako bi favorizovao njegov i pro-republikanske naloge na toj mreži, što je rezultiralo znacajnim porastom pregleda, lajkova i dijeljenja sadržaja sa tih naloga.
Koga zanima detaljnije: eprints.qut.edu.au/253211/1/A_c...
Koga zanima detaljnije: eprints.qut.edu.au/253211/1/A_c...
November 16, 2024 at 8:06 PM
Izgleda da je Musk sredinom godine prilagodio X algoritam kako bi favorizovao njegov i pro-republikanske naloge na toj mreži, što je rezultiralo znacajnim porastom pregleda, lajkova i dijeljenja sadržaja sa tih naloga.
Koga zanima detaljnije: eprints.qut.edu.au/253211/1/A_c...
Koga zanima detaljnije: eprints.qut.edu.au/253211/1/A_c...
Then we solve for a and take the limit as q goes to 1 (which would give us perfect overlap) which gives us our critical capacity
We have our critical value a_c = p/N where p is number of patterns, N number of neurons, at which the volume of valid solution space shrinks to 0!
We are done!🎉 (62/62)
We have our critical value a_c = p/N where p is number of patterns, N number of neurons, at which the volume of valid solution space shrinks to 0!
We are done!🎉 (62/62)
![$$\boxed{a_c = \left[\int_{-\kappa}^\infty Dt(t + \kappa)^2\right]^{-1}}$$](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:4mx5fgnf6nx6hlfngtfivh3y/bafkreiehsksiolkyn32wq3doekycqkn4h6g4tph5h2nnh45nobbcgwh6lm)
March 19, 2026 at 9:49 AM
Then we solve for a and take the limit as q goes to 1 (which would give us perfect overlap) which gives us our critical capacity
We have our critical value a_c = p/N where p is number of patterns, N number of neurons, at which the volume of valid solution space shrinks to 0!
We are done!🎉 (62/62)
We have our critical value a_c = p/N where p is number of patterns, N number of neurons, at which the volume of valid solution space shrinks to 0!
We are done!🎉 (62/62)
Per the image,
(and knowing that m/s = km/h / 3.6)
v = 1.7 m/s
r = 2 m
A_c = 1.4 m/s ~ 0.15 G
v = 28 m/s
r = 15 m
A_c = 52 m/s ~ 5.3 G 🤔 O RLY?
On checking - top speed is about 45 km/h, not 100 km/h; then
v = 12.5 m/s
r = 15 m
A_c = 10.4 m/s ~ 1.05 G 👈 more like it.
2/
(and knowing that m/s = km/h / 3.6)
v = 1.7 m/s
r = 2 m
A_c = 1.4 m/s ~ 0.15 G
v = 28 m/s
r = 15 m
A_c = 52 m/s ~ 5.3 G 🤔 O RLY?
On checking - top speed is about 45 km/h, not 100 km/h; then
v = 12.5 m/s
r = 15 m
A_c = 10.4 m/s ~ 1.05 G 👈 more like it.
2/
October 18, 2023 at 5:31 PM
Per the image,
(and knowing that m/s = km/h / 3.6)
v = 1.7 m/s
r = 2 m
A_c = 1.4 m/s ~ 0.15 G
v = 28 m/s
r = 15 m
A_c = 52 m/s ~ 5.3 G 🤔 O RLY?
On checking - top speed is about 45 km/h, not 100 km/h; then
v = 12.5 m/s
r = 15 m
A_c = 10.4 m/s ~ 1.05 G 👈 more like it.
2/
(and knowing that m/s = km/h / 3.6)
v = 1.7 m/s
r = 2 m
A_c = 1.4 m/s ~ 0.15 G
v = 28 m/s
r = 15 m
A_c = 52 m/s ~ 5.3 G 🤔 O RLY?
On checking - top speed is about 45 km/h, not 100 km/h; then
v = 12.5 m/s
r = 15 m
A_c = 10.4 m/s ~ 1.05 G 👈 more like it.
2/
Seems to align with the findings from a study done by researchers at Queensland University of Tech in Australia. That study showed that the X algo may have been altered on Nov 12/13th amplifying Musk’s and other top right wing accounts. Right before the election.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au
January 23, 2025 at 11:36 PM
Seems to align with the findings from a study done by researchers at Queensland University of Tech in Australia. That study showed that the X algo may have been altered on Nov 12/13th amplifying Musk’s and other top right wing accounts. Right before the election.
eprints.qut.edu.au/253211/1/A_c...
eprints.qut.edu.au/253211/1/A_c...