At Toba aquarium in Japan, after closing time, some clever little otter pups help their grandpa tidy up their toys. As a reward, he gives them ice cubes
I saw the photo in my feed and went ohh, dude, no, we do not handle yellow rocks with our bare hands until we know for sure what they are. And I know that orange...
In comments, they continue:
and that's where I started cussing at the computer monitor. But someone else had got there first:
So just as a reminder, folks. If you don't know what it is, don't put it in your fucking mouth!
when will they establish that "james bond" is like an intergalactic slug that possesses random british men every 5 to 15 years and uses its collected life experiences to go on spy adventures
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.
Rumination is probably the most common type of OCD compulsion, but I rarely see anyone talking about it. I've talked to multiple people diagnosed with OCD who didn't even recognize it as a compulsion.
Basically, if you have OCD you have terrible intrusive thoughts. They can be about anything, but common themes are fear of being a bad person, fear of hurting someone, fear of contamination. etc.
Rumination is when you get stuck in a spiral. Rumination is when you spend hours catastrophizing, overthinking, analyzing, telling yourself it's going to be okay.
I'll say it again:
Rumination is a compulsion.
Rumination is a compulsion, and that means you have to stop doing it.
I did ERP (exposure response prevention) for my OCD with a therapist! For 9 months! And it did help, but the idea didn't really click until I found this website a couple years later.
And Oh My God. It made things make so much more sense, and I was able to pull myself out of an episode even though I wasn't in therapy or on meds at the time.
Genuinely if you have OCD, or even if you suspect you have OCD, I'm begging you to read some of these articles.
Like this was genuinely life changing for me.
Here are some of the ones that were most helpful to me:
One thing I really like about Beverly Engel's book It Wasn't Your Fault, which is about PTSD-induced toxic shame, is that quite a bit of it deals with people who haven't broken The Cycle of Abuse (TM) and have gone on to hurt others. That's a really underserved and vulnerable patient population, and statistically, it's also MASSIVE. I don't think I've read a single other self-help type book on PTSD and self-loathing that confronts the possibility that you're exactly as bad as you think you are.
I felt better that it so much as mentioned that children can react to abuse with ungovernable rage. Everybody likes the image of PTSD patients as internalizing everything and becoming doormats, which does happen, and often, but it's not the only narrative. Personally I've always hated my abusers and have always wanted everyone who so much as breathed wrong in my direction from ages 0 to 18 to burn eternally in hell. I *never* thought any of it was my fault and ever since I was a toddler I was willing to make it everybody else's problem, and it's really relieving to read a clinical perspective that acknowledges that abuse victims can act that way too.
It's wild to me that its such a neglected subset of abuse victims. Its really common. When I still lived with my parents and was still subjected to my father every fucking day I would lash out terribly at my mother, to the point when i went to visit them for years afterwards she was afraid I would lash out again. We've worked it out, I'm a much better person when I'm not regularly subjected to mental and emotional abuse, but like, its just so common.
I think it must be, at least partially, because, people hate the imperfect victim. Its easy for so many people to sympathize with someone who never lashed out. Less so for people to sympathize with people who are angry and lash out. Even though its a perfectly sensible reaction to being hurt over and over. I'm sure most people would like to think they would simply never.
I don't think this is the whole reason, but, I think it plays into it.
Similarly, there’s a narrative of, you cannot experience grief over having fucked up. That if you are hurting because you caused harm, because you were the cause of harm, that you’re not allowed to grieve, because you “earned your sorrow. You deserve to bottle it up and to hurt for the bad things you have done,”
Which is punitive logic. It’s copthink. Which is bad.
In fact, you can actually give yourself trauma over fucking up too badly and doing, witnessing, or failing to prevent something evil that goes against your morals; for instance, if you steal your mother's life savings due to drug addiction, kill a civilian during a military operation (please do not join the military), or became abusive because you didn't have the tools and skills yet to handle BPD. In the field of psychology this is called "moral injury" or "perpetrator trauma."
ellievsbear
origamiaround
kianamaiart
blakekathryn
vaporwave
discoholicmusic
stormborns
andulkaphoto
titsay
