June 1 – Today is the first day of LGBTQIA+ Pride Month. A few days ago, I organized a vigil for a neighbor who was murdered. Eryka Caldwel
By Melinda Butterfield
Across the U.S., at least 12 trans people have been murdered or found dead in suspicious circumstances since early March. That’s about one per week.
On the surface, Eryka Caldwell and Juniper Blessing don’t seem to have had much in common: from different generations, different backgrounds, living very different lives on opposite ends of the country. But what they did have in common was being members of a marginalized group that has been relentlessly scapegoated by the wealthy and powerful for the last half-decade.
The back-to-back murders of trans women in Seattle and Brooklyn, havens of trans life and queer culture, have made explicit what we already knew in our guts: in 2026, there is no safe place for trans people in the United States.
The worst thing you can do, as someone who has recently realised they are transfem, is to let terves and transphobes convince you cis women will never accept you.
I was told that when I came out everyone would reject me. That I would find myself isolated from the world, and from other women especially, who would react to me with horror and revulsion.
In reality, within the first months of coming out, in no particular order:
My sister's reaction on my coming out was, "Right, so I have a sister instead of a brother. Cool. I'm taking you clothes shopping tomorrow."
A friend, when she learned I am a woman, immediately invited me to her women-only, girls-night-out birthday party the following week.
Another friend, when a friend of hers expressed doubts about my gender, immediately shut them down and reaffirmed I am a woman.
I went camping with a group of friends, and we had two tents, one for the boys and one for the girls; I was unsure as to which I should enter, to which a girl friend responded by grabbing me and physically dragging me inside the women's tent.
In the women's bathroom at a movie theatre a random woman, whom I'd never seen before and haven't seen since, stopped me as I was going into a stall, to warn me there was no toilet paper in there, because she'd just used the last of it.
All of these, and more, some from friends, some from complete strangers. All within a few months, as a trans woman who hadn't started medical transition yet, and was very visible as being a trans woman.
I've had some people reject me, true, but the vast majority, including almost all cis women, accepted me as a sister with open arms.
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.
A computer distribution program says it’s seeing a big increase in the number of laptops being donated and refurbished to students across Nu
In an overcrowded office, Joseph Kanayuk-Driscoll pulls out a screwdriver and begins to dismantle a laptop.
It’s one of dozens he’s worked on this week as a computer refurbisher in the Ampere office in Iqaluit. He’s part of a computer distribution program that takes old donated laptops and repairs them to give to students and people across Nunavut in an effort to keep e-waste out of Iqaluit landfills.
“What excites me is actually being able to give lots out to the community and making sure the laptops that are completely available to use and still work keep out of the dump,” he said.
reblogged a post yesterday talking about how hot consent is and hearing someone say “please” and I feel like I would like to add how how it also is when someone is thanking you for letting them touch you.😵💫
Eryka Caldwell is a back trans woman who was murdered by her partner in her apartment last Sunday, and the story is getting a fraction of the attention that the murder of Juniper Blessing did. The police had already been called several times about her partners violence before her murder and did nothing, and she deserves the same outrage and mourning as Juniper got, and every one of our murdered trans siblings deserves. Trans women of color are more likely to be the victims of murder than any other group of queer people, and they need our solidarity, protection and support.
Caldwell’s boyfriend, 38-year-old Jonathan Fernandez, has been charged with murder.
Three things have happened in my local trans community in the past month:
A transmasc drag queen made a call-out post about how a newly-out local trans woman is a sexual predator. (The sexual offence was that, 6 years ago, before she came out, she was at a drag show with the drag queen (who was her friend at the time) and joked that the drag queen's makeup looked a bit like semen.) The trans woman spiralled, got drunk, got hit by a car, and is currently in hospital.
A newly-out trans woman and a trans man met at a local trans support group. They decided to meet up again at a local pub. She apparently asked questions about his transition that he considered invasive, and in a voice that was too loud and risked outing him to other people in the venue. She now faces being banned from the support group.
The only trans woman who volunteers to run that support group is being kicked off the team by the trans man who is in charge of it. The reason is that she once told the guy in charge to "stop talking" when he was giving her several instructions at once, and also that a different transmasc volunteer has said they find her "creepy" and don't want to be alone with her.
All three trans women are autistic, isolated, and don't have much practice with irl socialising. Maybe they're a little bit loud, maybe a little bit abrasive, maybe they misjudge the appropriateness of sexual comment to a friend at a drag show, maybe they're too keen to make a new friend that they get excitable and speak too loud and ask too many questions, maybe they get overstimulated by being told too many instructions and need you to stop for a second to give them space.
Maybe their behaviour that you consider weird and unsettling is actually a trauma response to a childhood and adolescence of isolation and bullying.
And you know what the funny thing is? I have been sexually harassed at a drag show by the drag queen in (1), who was performing and making sexual jokes about various people in the audience. I have been in a public place with the trans man in (2) while he asked me loud invasive questions about my HRT regimen and about child abuse I suffered. And I have been misgendered and deadnamed repeatedly by the trans man in (3).
And do you know what I did about any of these things? Nothing, beyond saying "Hey, I'd appreciate if you don't do that next time 👍 No hard feelings 🙂". Because, not only am I aware that people make mistakes, and I consider trans men and transmascs a vulnerable demographic and it would be a totally unjust overreaction from me to attempt to socially isolate them for these offences.
But also, I don't have the social power to do anything about these things anyway, even if I wanted to. My normal response to experiencing even serious abuse at a queer event is to simply stop attending it, isolating myself, and perhaps telling a few close friends about it. Because I already know that my complaints will not be taken seriously.
Trans women are being isolated irl every day for behaviours that other demographics can do with impunity. They are not given any grace, any benefit of the doubt, any second chances, and complaints about even the smallest offence will be taken seriously and escalated. And they will not be able to fight back, because a trans woman who fights back is just confirming all the bigoted assumptions that the wider world already has about her.
white people really cant give a shit abt anyone but themselves unless held at gunpoint or smth huh?
“what about the poor little white kids just trying to have fun?” what about the poor little kids of color made to feel unsafe and unwelcome in a space that was supposed to be fun?
“why can’t poc just make their own redesigns and we can keep these ones” why cant you realize that your actions have consequences and that you should maybe try to not enforce racist stereotypes??
like ofc its ideal to have VARIETY among both villains AND protagonists, its when your characters look like this
that the issue arises
scratch “can”, white people are encouraged to reblog
black screen. slow jazz saxophone. dramatic close-up slide along the side of the car interrupted by a tail swishing through the shot. when the camera reaches the hood, it zooms out and pans upwards to reveal the front half of a dragon sprawled out on top of the car, eyes rolling back and long forked tongue lolling out, obviously immediately post-orgasm
sultry female voice: a dragon wouldn’t choose anything else, so why should you? buy a subaru
