Hi hello! Welcome to my Just A Regular ol' Blog. Call me elio (or any of the other names you know me by). Any pronouns are ok. I'm not consistent about anything so if you follow me I guess it's because we share some obsessions, or you're interested in my ramblings (wow are we best friends now?)
A few of the repeating themes of my blog:
STEM posts, more often engineering and software. Because yeah
Politics
Self care :)
Aroace awareness, ramblings, etc
My own writing (rare)
Fandom stuff because really that's the main reason I'm even on this website
Ok here's a non-exhaustive list of my fandoms!
Project Hail Mary
Alien Stage
To Be Hero X
Omniscient Reader's Viewpoint
Genshin Impact and Honkai Star Rail
Little Mushroom 🍄🤍❤️
Be warned: I'm TERRIBLE at keeping consistent tags. Especially for older posts. Unless I tag something with a fandom tag (e.g. #tbhx, #alnst, or otherwise), then it can be literally anything else.
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.
Game companies hate emulation, but none of them seem to understand that a lot of us would just buy ROMs from them directly if we could. I don't want a fifth remake of Final Fantasy IV, I want to pay five bucks for the 3MB file you already made bank with thirty years ago. Nobody who wants to play something for the purpose of retro gaming is going to consider a $40 remake as the alternative option, and we're certainly not going to let the original dissappear. They're crying about opportunity cost for a product they're not even selling.
op i know you're probably talking about like, video games, etc, but this is also critical for research science - my lab has so much abandonware, either because the company's out of business, or the company decided to not maintain it, and it's a fucking nightmare. we have two windows 95 computers that are CRITICAL for performing experiments/data analysis because the software needed is abandonware. one of the main roles for a guy in my lab is to maintain these little dinosaurs because if they go out, we lose access to ~20 years of raw data for research. part of why is that these companies also make their own file types, and make it difficult-to-impossible to convert those file types without their specific software. by habit, i convert all research files to more generic versions (txt, pdf, tif, etc) so that i minimize risk of losing my shit, but some stuff can't be converted.
for example, we have a microscope that is perfectly functional, good microscope, but its software is abandonware because the company refused to maintain it. the company is still in business, still makes essentially the exact same software, but they made all of the old tech incompatible with new software to force people to buy the new microscope tech. it would cost a quarter million dollars to replace this microscope. this perfectly good microscope.
so like, i know a lot of people look at the original post here and go "well op just wants old video games to play" (which is valid! games companies should not be able to push shit to abandonware and then close it off) but also this is critical for like. biomedical research. if y'all had any idea how much basic infrastructure built on science relies on shit that is technically abandonware, you would probably be horrified.
Everyone knows the first day of Friend Grace’s class is nickname day. It’s the day when every pebble is on their best behavior to try and make sure they get a cool nickname, something unique that they can brag to their friends and classmates about.
Sometimes, Grace will do it without thinking. That’s how Kiddo and Buddy got their nicknames. Often, Grace will nickname students after their coloration. Gaia got his nickname because he’s blue and green, and apparently looks a lot like Earth. Violet got hers because she’s purple. (She was initially disappointed since color means nothing to Eridians, but then Friend Grace showed them violet flowers and said that humans often associated purple with wealth and royalty, and she changed her tune.) Most of the time, Grace will give his students what he calls “regular human names” like Abby, Carl, or Martin.
But the most coveted nicknames are ones named after Earthen creatures. When ♩♪♬ 🎵 ♩♪♬ 🎵 first introduced themselves, Friend Grace immediately perked up and shouted “Robin!” After a bit of explaining himself and a few videos of bird calls, Robin was trilling and chirping happily, excited at having a nickname that felt like a 1-to-1 translation of their own.
Even well after Friend Grace is gone, his legacy remains. A hundred years into the future, when humankind finally launches a new ship with the express purpose of properly meeting their Eridian neighbors, one of the first messages exchanged is “Hello! My name Robin.”
......suddenly struck by the idea for a piece of worldbuilding of "fae don't like iron bc it is the most stable element*"
*as in elements higher you can extract energy via fission and lower you can extract energy via fusion but iron itself there is no excess binding energy to extract at all
Just found this but couldn’t reblog it. I had just finished organizing a doc for a fanfic that I’m working on with a friend…. I guess I need to fucking move it now. Ffs.