ultratumba

☿ rainbows & monsters || they/them ☿ 🪩🌈🫒🪭🏺🪞🐚🪸🌊

[r2x2]

The Amnesty International report that JK Rowling is trying to censor through legal threats.

It can also be found on Google Drive here and on Bluesky here.

[cbarc]

Please save and share this info where possible. This report has since been removed by Amnesty International due to legal threats of libel from JK Rowling. This information is no longer easily accessible.

[trannydurden]

happy juneteenth but dont forget that prisoners are legally allowed to be subject to slave labor and also black people are disproportionately arrested and subjected to that legality. happy juneteenth but slavery still lives in america. america is still dependant on slave labor.

[dogdykesdoitbetter]

"Over 4,100 corporations profit from mass incarceration in the United States."

"Around 63,000 inmates produce goods for external sale...from farm work and manufacturing to call center and distribution services."

"In Oklahoma and Mississippi, [work release] is not restricted to inmates. Oklahoma uses defendants sent to rehabilitation programs to process chickens at a private facility for no pay... Mississippi has ‘restitution programs’ where it sends people to work off debts owed to the state by acting as a temp agency for a number of industries."

"...federal immigration detention centers rely on the labor of civil immigrant detainees to lower their operating costs. The large prison corporations running the centers – like CoreCivic and GEO Group – are currently facing lawsuits, alleging that they employ deprivation schemes, where detained immigrants are forced to choose between living without basic necessities or working for sub-minimum wages, and then are threatened with punishment for refusing to work."

Private Companies Producing with US Prison Labor in 2020: Prison Labor in the US, Part II — Corporate Accountability Lab

This post follows on from our post exploring the use of prison labor during the Covid-19 pandemic. It details the current forms that priso

[akonoadham]

Footage of Juneteenth parades and events from 1979-1991 broadcasted on the KXAS-TV station in Fort Worth, Texas.

[uncleromeo]

if you don't do anything else today,

Please have a moment of silence for the people who were killed instead of freed when news of emancipation finally reached the furthest corners of the american south.

have another moment for the ledgers, catalogs, and records that were burned and the homes that were destroyed to hide the presence of very much alive and still enslaved people on dozens of plantations and homesteads across the south for decades after emancipation.

and have a third moment for those who were hunted and killed while fleeing the south to find safety across the border, overseas, in the north and to the west.

black people. light a candle, write a note to those who have passed telling them what you have achieved in spite of the racist and intolerant conditions of this world, feel the warmth of the flame under your hand, say a prayer of rememberance if you are religious, place the note under the candle, and then blow it out.

if you have children, sit them down and tell them anything you know about the life of oldest black person you've ever met. it doesn't have to be your own family. tell them what you know about what life was like for us in the days, years, decades after emancipation. if you don't know much, look it up and learn about it together.

This is Juneteenth.

white people CAN interact with this post. share it, spread it.

[alwaysbewoke]

—Did you know that Juneteenth is also celebrated in a part of Mexico? Nacimiento Mexico was once home to thousands who escaped slavery in the US. As many as 10,000 slaves followed a clandestine Southern Underground Railroad to Mexico. —To date, many Black Mexicans from the Texas area retrace a portion of the same route their African American ancestors followed in 1850 when they escaped slavery. —Descendants of slaves who escaped across the southern border observe Texas’s emancipation holiday with their own unique traditions in the village of Nacimiento. —Slave hunters would patrol the southern border for escapees, led by the Texas Rangers but the Mexican army would be there waiting for them (the slave hunters) to turn them away.

x

[transitoryparentheses]

This NPR Article!!

A Chapter In U.S. History Often Ignored: The Flight Of Runaway Slaves To Mexico

As the U.S. Treasury considers putting Harriet Tubman on the $20 bill to honor her role in the northbound underground railroad, new attentio

[lelouch]

Also, in the Mexican constitution, the first article mentions how slavery is prohibited in México and every enslaved person arriving here would be freed by virtue of being on Mexican soil, and protected by our laws

[homoqueerjewhobbit]

The crowd was as much a part of the show as the band. Everyone was there: North Beach beatniks and barrio zoots, the bored bikers in black leather, teenagers in the back row kissing. There were long-haired, lithe girls in belly-dance get-ups, pink-haired punks safety-pinned together, hippie suburbanites, movie stars so beautiful they left you dumbstruck, muscle gayboys with perfect mustaches, butch dykes in blue jeans, and fairies of all genders in thrift-store dresses. We rode the mirrored ball on glittering LSD and love power. Dance fused us, magical and cleansing. We were all in a swirl of color and light. It was like a rainbow.

A rainbow. That’s the moment when I knew exactly what kind of flag I would make.

-Excerpt from Gilbert Baker's memoirs.

[thedouble]

“Freedom always has a price.”

― Persepolis (2007) dir. Marjane Satrapi, Vincent Paronnaud

[aeli-noir]

[justalurkr]

Text of tweet under the cut because it is loooong.

But... Stochastic Parrots.

Timnit Gebru was fired from Google in December 2020 for refusing to retract a research paper, and every single warning that paper made about large language models has now happened at a scale the industry spent 4 years trying to make people forget about.

Her name is Timnit Gebru.

She co-led the Ethical AI team at Google. She co-wrote a paper called "On the Dangers of Stochastic Parrots" with Emily Bender at the University of Washington and two other researchers. The paper was 14 pages long. It was submitted to a top AI ethics conference. And it was the reason Google decided that one of the most senior Black women in AI research could no longer work there.

The story Google told publicly was that she resigned. The story she told, confirmed by 2,695 of her colleagues in an open letter, was that she was fired by email while on vacation because she refused to either retract the paper or remove her name from it.

The paper had not even been published yet.

Here is what she actually wrote, and why every prediction inside it has now come true.

The first warning was about scale itself. Bender and Gebru argued that training ever-larger models on ever-larger scrapes of the internet would produce systems that appeared fluent but had no actual understanding of language. They called these systems stochastic parrots because they would repeat patterns from training data with statistical confidence and zero comprehension. The paper predicted that this apparent intelligence would fool both users and developers into trusting outputs that were structurally incapable of being reliable.

This was 2020. GPT-3 had just come out. The paper predicted the hallucination problem before anyone had a word for it.

The second warning was about bias amplification. The paper documented in detail that internet-scale training data contains systematic overrepresentation of dominant viewpoints and underrepresentation of marginalized ones. The models would not just absorb this bias. They would amplify it, because the optimization process rewards confident outputs, and confidence in language patterns tracks frequency in the training set.

The prediction was that hiring tools built on these models would discriminate against women. That healthcare triage tools would underperform on Black patients. That loan approval systems would entrench inequality while presenting their decisions as neutral algorithmic judgment.

Every one of those things has now been documented in deployment.

Amazon's hiring algorithm penalized resumes that contained the word "women" in any context. Healthcare risk scoring algorithms used by major US hospitals were found to systematically underestimate the medical needs of Black patients. Apple Card's credit algorithm gave wives credit lines 10x lower than their husbands for the same financial profile.

The third warning was about environmental cost. The paper calculated that training a single large language model produced emissions equivalent to the lifetime output of 5 cars. The prediction was that the race to scale would create an environmental footprint that would eventually rival entire industries.

In 2024, Google's emissions were up 48% from 2019, and the company explicitly blamed AI infrastructure. Microsoft's were up 29%, same reason. Both companies have now quietly abandoned the climate commitments they were publicly celebrating the year Gebru was fired.

The fourth warning was about documentation. The paper argued that the training datasets being assembled were too large for anyone to actually audit. Nobody at Google, OpenAI, Meta, or any other lab could tell you with confidence what was in the data their models were trained on. This was not a temporary problem to be solved later. It was a permanent feature of the approach.

In 2023, researchers discovered that the LAION-5B dataset, used to train Stable Diffusion and other major image models, contained thousands of images of child sexual abuse material. The companies that had trained on the dataset had no way of knowing. The paper predicted that category of failure 3 years before it was found.

The fifth warning was the one Google cared about most.

Bender and Gebru argued that the deployment of these systems would centralize linguistic and cultural power in the hands of the small number of companies that could afford to train them. The internet would become a place where the dominant voice was a statistical average of dominant voices, presented as a neutral assistant. Languages underrepresented in the training data would degrade over time as more web content was generated by these systems and fed back into the next training run.

This is now happening in real time. A 2024 study found that 57% of new web content in English is AI-generated or AI-assisted. Researchers studying low-resource languages have documented active degradation in translation quality, because the synthetic content fed back into training is itself worse in those languages.

The paper Google fired her for predicted the model collapse problem before model collapse had a name.

The mechanism behind why this all happened is the part of her work that nobody quotes.

Gebru's argument was not that AI is dangerous in some abstract sci-fi sense. Her argument was that AI is dangerous in a very specific structural sense. The technology was being built by a small group of researchers who shared similar backgrounds, worked at similar companies, and were rewarded for shipping products faster than competitors. The incentive structure made it impossible for safety, ethics, and bias concerns to slow anything down. Anyone inside the system who raised those concerns was either ignored, sidelined, or removed.

She was making that argument from inside Google.

Then Google proved her right by removing her.

The team Google had built to make sure their AI was safe was dismantled in 90 days because they did the job they had been hired to do. Margaret Mitchell, the other co-lead of the Ethical AI team, was fired two months after Gebru for searching through her own emails for evidence of how Gebru had been treated.

Gebru did not stop. She founded DAIR, the Distributed AI Research Institute, in 2021. The mission is to do AI research outside the control of the companies that have a financial interest in not hearing the answers.

Every prediction in the Stochastic Parrots paper has now been validated by deployment. Hallucinations are an industry-wide problem the largest labs cannot solve. Bias amplification has been documented in hiring, healthcare, lending, and criminal justice. Environmental costs are larger than entire small countries. Training data audits remain impossible. Model collapse is an active research crisis at every major lab.

The question worth sitting with is the one almost no one in the industry will say out loud.

Every researcher with the technical credibility to call out these problems watched what happened to her in December 2020 and made a calculation about their own career. The number of people willing to speak publicly about safety and ethics issues inside the major AI labs collapsed after that firing and has not recovered.

The researcher Google fired for warning about exactly what is now happening was right.

The company that fired her is now the second-largest deployer of the technology she warned about.

And the people inside that company who agree with her are not allowed to say so.

[medaknight4]

Here's a link to the paper

And a link to the Wikipedia entry for the phenomenon the paper covers

[lorquiana]

40 days. One email. Your name on the right side of history.

the new ehrc guidance re: the supreme court ruling on the equality act was released on may 21st after consultation. it is not good. you can read transactual's statement on it here. parliament has until june 30th to block this. for those of you in the uk, the link above is a tool you can use to email your mp quickly and ask them to do so. the website will talk you through the process but all you need to do is find your mp's address and personalise the sections in brackets, and this will do the rest.

if you are able, please email. it will take a couple minutes maximum.

[lorquiana]

Petition: Review anti-discrimination laws to strengthen trans women & LGBTQ+ protections

Launch a review into strengthening legal protections and clearer enforcement against discrimination, harassment and exclusion of trans women

there is a petition that uk folks can sign also - currently at 2,000 signatures out of 10,000 needed for a government response

[395nm]

Fuck Meyer-Briggs whatever typology. This INTFP shit is only for redditors up their own asses to substitute for a personality. Use my new typology instead!

Your ideal environment is:

Hot/Cold

Wet/Dry

Bright/Dark

Loud/Quiet

[bedupolker]

HWBL - beach boy

HWBQ - tropical fish

HWDL - dingy club bathroom hookup

HWDQ - the swamp woman

HDBL - CoachellaBurningmanSouthbysouthwestACL attendee

HDBQ - Lizard

HDDL - Vegas babeyyyy

HDDQ - Trapped in a slot canyon

CWBL - Rowdy Lobsterman Crew

CWBQ - penguin living

CWDL - port angeles basement show

CWDQ - bruminating amphibian/hypothermic mammal

CDBL - ski resort

CDBQ - Christmas in Nebraska

CDDL - mcmurdo station rave

CDDQ - corpse

[yesterdaysprint]

The Coffeyville Daily Journal, Kansas, April 30, 1896

[t3acupz]

before neil banging out the tunes there was czar scraping the strings

[therizinosaurus--cheloniformis]

Happy 20th Anniversary of Neil Banging Out The Tunes!!!!!

More rarer images of Neil, my beloved:

20 years.

I find it so beautiful that this little rat's life has been remembered with love for 20 years and will be hopefully for decades to come. Truly a marvel of the internet.

<3 love you Neil!!!!

[sobcat]

THAT'S US!! :D ON THE RIGHT

[ateazure]

"and the stars look very different today..."

[monidoll]

its kinda hard to celebrate trans day of visibility because the visibility we're getting lately is like

[yet-another-heathen]

[yet-another-heathen]

[tpwrtrmnky]

Trans day of every tabloid shuts the fuck up about us forever. Trans day of let us have our healthcare and leave us the fuck alone. Trans day of tearing down the panopticon. Trans day of let us control the narrative instead of deferring to some cis sexologist's hallucination.