alright I've got to do some quick math to explain attitudes towards AI to my boss.
we're looking to create an AI policy, and when we were talking about this, my boss (older millennial) was genuinely shocked to hear that younger people do not (seem) to view AI positively (a la the recent commencement speakers being booed)
please rb for larger sample size!
Question 1/3
What is your age, and do you feel AI is a net positive or net negative in our lives today?
This is how the system of white supremacy operates. The media is used 2 create stereotypes like blk on blk crime.They need black men to fill jail cells for the Prison Indstrial complex
You know what? I’m tired of this.
I do not know what exactly they are waiting for. I mean our government comes up with “reasons” to invade other countries, such as Syria, like their government is allegedly violating human rights or something like that. but… I mean for other countries, they do not even have to go deep to bomb the fuck out of this place, they can just look at our media. And this has been happening to people of color since the media has existed.
Did a research project on this in undergrad and the results are extremely alarming because it’s not just in imagery, it’s in language used even in the law making process and within our own communities in a completely different way than expected.
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.
According to the CDC, in 10 percent of those drownings, the adult will actually watch the child do it, having no idea it is happening. Drowning does not look like drowning—Dr. Pia, in an article in the Coast Guard’s On Scene magazine, described the Instinctive Drowning Response like this:
“Except in rare circumstances, drowning people are physiologically unable to call out for help. The respiratory system was designed for breathing. Speech is the secondary or overlaid function. Breathing must be fulfilled before speech occurs.
Drowning people’s mouths alternately sink below and reappear above the surface of the water. The mouths of drowning people are not above the surface of the water long enough for them to exhale, inhale, and call out for help. When the drowning people’s mouths are above the surface, they exhale and inhale quickly as their mouths start to sink below the surface of the water.
Drowning people cannot wave for help. Nature instinctively forces them to extend their arms laterally and press down on the water’s surface. Pressing down on the surface of the water permits drowning people to leverage their bodies so they can lift their mouths out of the water to breathe.
Throughout the Instinctive Drowning Response, drowning people cannot voluntarily control their arm movements. Physiologically, drowning people who are struggling on the surface of the water cannot stop drowning and perform voluntary movements such as waving for help, moving toward a rescuer, or reaching out for a piece of rescue equipment.
From beginning to end of the Instinctive Drowning Response people’s bodies remain upright in the water, with no evidence of a supporting kick. Unless rescued by a trained lifeguard, these drowning people can only struggle on the surface of the water from 20 to 60 seconds before submersion occurs.”
This doesn’t mean that a person that is yelling for help and thrashing isn’t in real trouble—they are experiencing aquatic distress. Not always present before the Instinctive Drowning Response, aquatic distress doesn’t last long—but unlike true drowning, these victims can still assist in their own rescue. They can grab lifelines, throw rings, etc.
Look for these other signs of drowning when persons are in the water:
Head low in the water, mouth at water level
Head tilted back with mouth open
Eyes glassy and empty, unable to focus
Eyes closed
Hair over forehead or eyes
Not using legs—vertical
Hyperventilating or gasping
Trying to swim in a particular direction but not making headway
Trying to roll over on the back
Appear to be climbing an invisible ladder
So if a crew member falls overboard and everything looks OK—don’t be too sure. Sometimes the most common indication that someone is drowning is that they don’t look like they’re drowning. They may just look like they are treading water and looking up at the deck. One way to be sure? Ask them, “Are you all right?” If they can answer at all—they probably are. If they return a blank stare, you may have less than 30 seconds to get to them. And parents—children playing in the water make noise. When they get quiet, you get to them and find out why.
Can I just say thank you to OP for putting such a detailed description on this?
I’ve been a lifeguard for 6 years now and of all the saves I’ve done, maybe two or three had people drowning in the stereotypical thrashing style. And even those, like the save I made last weekend, it was exactly like OP describes where the person’s head is going in and out of the water but it isn’t long enough to get any air. Mostly you recognize drowning by the look on someone’s face. If someone looks wide eyed and terrified or confused, chances are they’re drowning. That look of “oh shit” is pretty easily recognizable. And even if you can’t tell for sure: GO AFTER THEM ANYWAY. I’ve done “saves” where a kid was pretending to drown and I mistook it for real drowning, but that’s preferable to a kid ACTUALLY drowning.
Also please remember that even strong swimmers can drown if they have a medical emergency, get cramps, or get too tired. If your friend knows how to swim but they’re acting funny get them to land. And even if someone can respond when you ask them if they need help, if they say they do need help? GO HELP THEM.
However . If the victim is a stranger, I can’t recommend trying to get them. Lifeguards literally train to escape “attacks,” because people who are drowning can freak the fuck out and grab you and make YOU drown as well. If you do go in after someone, take hold of them from the back and talk to them the whole time. IF YOU ARE GRABBED: duck down into the water as low as you can get. The person is panicking and won’t want to go under water and should release you. Shove up at their hands and push them away from you as you duck under. Don’t die trying to save someone else.
Please guys, read and memorize this post. Not all places have lifeguards. Being able to recognize drowning is such an important skill to have and you can save someone’s life.
In a water park once, I was suddenly grabbed by a child and he dragged me under the water without warning. I was going to get angry with him when I resurfaced because I thought he was being an ass, until I looked at him go back in and out hyperventilating the entire time. I grabbed him under his arms and began trying to drag him out while screaming for the lifeguard.
When the lifeguard got us both out, a woman came running down and accused me of harming him and said he had been completely fine in the water. That there was no reason to drag him out of there. The lifeguard had to explain to her that her son had been drowning, to which her response was to say that she didn’t hear him call for help.
One more thing…my young, healthy and apparently a strong swimmer coworker drowned in Ottawa River; he was swimming with friends. They wanted to know who could stay underwater holding breath the longest. He never surfaced. The body was found several hours later. Please don’t play this kind of games.
Reading through these articles by Al Jazeera and the HRW had my heart stop cold. Not only do military trials of Palestinians have a 96% conviction rate and prisons full of horrible, torturous conditions, but now there's this. Just vile. Please give the articles a read if you've the time. There's more out there as well.
If you are interested in helping a family survive and escape Gaza, please consider contributing to my friend Hamza's fundraiser, or supporting him other ways. His family has several young children, and I worry about them deeply.
Then you’re gonna love this photo of Annie Jump Canon.
Working at Harvard in the late 1800’s and early 1900’s as a “Computer”, Annie Jump Cannon cataloged stars using their spectra from photographic plates, in an effort to understand the mysteries and peculiarities of stellar spectra.
This was hard, detailed, nuanced work. By 1889, three years into her work, she had classified over 1,000 stars. By 1913, she could classify 200 stars an hour. She could classify three stars a minute, just by sight. Using a magnifying glass, she could classify stars down to 9th magnitude, 16 times fainter than the human eye can see. And she did this all with exceptional accuracy.
Over the course of her career, she personally classified more than 350,000 stars, accounting for a mind-boggling 98% of all contemporary stellar spectra classifications, a feat that wouldn’t be bested until the 1990’s with automated digital sky surveys.
Cannon used these classifications to develop the Harvard spectral classification system (O–B–A–F–G–K–M), organizing stars by surface temperature and physical properties.
It is hard to overstate just how foundational her work was to modern astronomy and astrophysics. Her classifications have enabled more than a century of breakthroughs in stellar structure and evolution, including the understanding of how stars change over time and how temperature, luminosity, and composition are related. The system underpins the Hertzsprung–Russell (HR) diagram, one of the most important tools in astrophysics, and remains embedded in modern research, from stellar population studies to galaxy evolution.
The immense scale of her work was itself a massive contribution to astronomy. For comparison, before Cannon, star catalogs contained between 600 and 4,000 stars. Her work single-handedly proved that large-scale stellar classification was both feasible and scientifically valuable. She helped establish systematic star catalogs as a core method of modern astronomy and laid the groundwork for astrophysical research on stellar structure, evolution, and populations that continues today.
Stop posting news as unsourced screenshots!! Please, I beg!!
Include the link to the source or it didn't happen!!!
And preferably include the source (website) name and date/year in the post too!!!
This has been a very loud, but very humble beseechment from your local media professional, thank you for your time
Seriously though, the spread of misinformation and disinformation is higher than ever, thanks to AI, botnets, 4chan, fascists, psyops, etc. Please use good sourcing!!! This is legit a way to help keep your communities safe.
If you see something and you don't know if it's true, or if the source is reliable, the two best/most comprehensive places to check are:
Snopes - to check if something is true/real, including social media posts and images/videos
Media Bias/Fact Check - to check if a website, news outlet, or other source is reliable. They rate outlets on accuracy of reporting and on political bias, and maintain detailed fact-checking records
I encourage you to check them whenever you're not sure about something important - the bigger and more upsetting it is, the more important it is to check! And I also encourage you to bookmark them for whenever are worried you might be reblogging something untrue or made by AI!
Stay safe out there, yall. Keep each other safe. By adding source links!!!
