1. The court holds Google responsible for statements made by its AI, considering them Google's statements (search engines have limited liability for results in their engine as they're the words of other sites/companies/people), meaning when their AI lies/hallucinates they're liable for the defamation/harm resulting from those statements.
2. Google's defense that customers are generally aware of the lack of reliability and are responsible for fact checking was dismissed. As the court pointed out, that would "significantly diminish" AI Search's stated purpose and it can't be distinguished from Google's business practices/statements as a search tool.
3. Studies have found about 91% of Google's everyday AI responses are accurate, leaving millions of searches per HOUR with potential liability for falsehoods. 56% of correct responses weren't supported by the sources the AI listed. Both of which mean Google is now liable for a LOT more AI "errors."
4. Google was held liable for 80% of court costs in this case and this precedent is expected to reverberate around the world. This is a massive shift from the 3rd-party search provider role Google has previously played and it comes right as they've tied ALL searches to their AI search.
The nazis that you see in movies are as much a historical fantasy as vikings with horned helmets and samurai cutting people in half.
The nazis were not some vague evil that wanted to hurt people for the sake of hurting them. They had specific goals which furthered a far right agenda, and they wanted to do harm to very specific groups, (largely slavs, jews, Romani, queer people, communists/leftists, and disabled people.)
The nazis didn't use soldiers in creepy gas masks as their main imagery that they sold to the german people, they used blond haired blue eyed families. Nor did they stand up on podiums saying that would wage an endless and brutal war, they gave speeches about protecting white Christian society from degenerates just like how conservatives do today.
Nazis weren't atheists or pagans. They were deeply Christian and Christianity was part of their ideology just like it is for modern conservatives. They spoke at lengths about defending their Christian nation from godless leftism. The ones who hated the catholic church hated it for protestant reasons. Nazi occultism was fringe within the party and never expected to become mainstream, and those occultists were still Christian, none of them ever claimed to be Satanists or Asatru.
Nazis were also not queer or disabled. They killed those groups, before they had a chance to kill almost anyone else actually. Despite the amount of disabled nazis or queer/queer coded nazis you'll see in movies and on TV, in reality they were very cishet and very able bodied. There was one high ranking nazi early on who was gay and the other nazis killed him for that. Saying the nazis were gay or disabled makes about as much sense as saying they were Jewish.
The nazis weren't mentally ill. As previously mentioned they hated disabled people, and this unquestionably included anyone neurodivergent. When the surviving nazi war criminals were given psychological tests after the war, they were shown to be some of the most neurotypical people out there.
The nazis weren't socialists. Full stop. They hated socialists. They got elected on hating socialists. They killed socialists. Hating all forms of lefitsm was a big part of their ideology, and especially a big part of how they sold themselves.
The nazis were not the supervillians you see on screen, not because they didn't do horrible things in real life, they most certainly did, but because they weren't that vague apolitical evil that exists for white American action heros to fight. They did horrible things because they had a right wing authoritarian political ideology, an ideology that is fundamentally the same as what most of the modern right wing believes.
Palestinians often smile while getting arrested as a symbol of resistance. They know they will face severe torture, isolation, discriminatory laws, and prejudice without fair trial yet smile to signify their strength.
âIf I have one message to give to the secular American people, itâs that the world is not divided into countries. The world is not divided between East and West. You are American, I am Iranian, we donât know each other, but we talk together and we understand each other perfectly. The difference between you and your government is much bigger than the difference between you and me. And the difference between me and my government is much bigger than the difference between me and you. And our governments are very much the same.â
To our dismay, the artist and author Marjane Satrapi has passed away.
As she wrote, ordinary Americans and Iranians have more in common with each other than we do with either of our governments, which have far too much in common with each other.
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.
For many years (ten now, about which, more soon) McMansion Hell has featured many prominent and diverse atrocities from all over these great United States and sometimes beyond them. However, most of these posts have consisted of houses built during the McMansion Era proper -- from the 80s up through around the early 2010s.
This is for a number of reasons. First of all: I like these houses because they are insane. Second of all, they are indeed quite different from one another -- they represent the owner's idiosyncratic if poorly rendered desires and fantasies. They are heavily psychologically loaded buildings. One family dreams endlessly of Tuscany, another wants to recreate the mall. All interiorize previously exterior forms of consumption.
These houses were also very expensive to build compared to their contemporary iterations: all real, solid wood cabinetry and trim, wrought iron railings, marble floors, elaborate murals - none of this is cheap. This is not to say that I'm nostalgic for the classical McMansion (though many are) only that it, like, most other facets of architectural and everyday life, have become progressively cheaper and more bland.
The McMansion never truly goes away. It merely changes shape over time. One of the shapes it currently takes is a particularly loathsome imitation of contemporary high architecture (specifically the kind of houses architects love to build for celebrities in California) executed in the most wretchedly parsimonious manner possible. It feels cheap to use the word 'slop' but their indiscriminate nature - the way they have no regard for why or how the things they imitate even work - allows it. Of all the building forms that could be generated with AI, this is the most likely. At any rate, behold:
Yes this is a real house. Yes you can buy it for $6 million in, yet again, Barrington, IL. It has 5 bedrooms and 5.5 bathrooms totaling 11,600 square feet. But most importantly, it looks like dogshit. Ten layers of Photoshop have been used to gussy it up which makes it appear entirely ersatz. Were it not for the interiors, I myself would have trouble trusting my own eyes. Part of the reason it looks so unreal is because the design itself is absurd, as though someone created four equally ugly vessels and threw them up one by one.
In 2017, in a now-deleted essay for Curbed (RIP - they destroyed the archive) I called these types of houses McModerns, simply because they were McMansions dressed up in modernist garb, which they wore no differently than they would Neo-Tudor or Mediterranean (broadly construed.) These houses don't warrant a new neologism, but they do feel like a degraded or perhaps even gonzo version of even that old concept. Slop works fine too, especially because half of what's in these images isn't real.
Much fascinates me about these houses, however one of the most unique elements vis a vis the last 30 years of building is how overtly and almost hostilely masculine they are. Anything that can be construed as feminized - color, softness, ornament - has been ruthlessly purged. They also rip off tech industry minimalism which only ads to their bro-ey nature. While previous iterations of McModernism (think new builds in Colorado with fake wood exteriors) scream dads with IPAs, these houses scream Reddit to me. They are Elon Musk-adjacent in sentiment.
By the way, this is what that room looks like without the fake furniture. It's basically a sunroom.
Whole Foods would like to call in a robbery.
Because these houses are designed by men, for men, no one involved has learned how a kitchen works. Many are calling this setup the "grindset tiktok video kitchen." This is the kitchen you see in those day in the life of an AI startup founder videos your algorithm forces you to watch against your will.
Virtual staging is actual literal slop. In fact, one can say that they were one of the first iterations of the ontological crisis we now face, one of the first instances where one is forced against one's will to question reality, what one sees with one's own eyes. Beyond that, I think virtual staging is literally a form of lying. You can use it to make a space look bigger or smaller than it is. In this it also has a lot in common with AI. This dining room has nothing to do with the world I'm living in. These chairs are not my problem.
It's actually AMAZING how much of what's in this house, beyond the furniture, is fake. Every single material is fake. The stone is aluminum paneling. The plants are plastic. The concrete is printed on some kind of surface (as evidenced through its repetitive pattern), though it's hard to say from just pictures. I don't even trust the floors!!
Ok if you haven't read Kelly Pendegrast's amazing essay "Merchandizing the Void" about how houses are all like stores now, HERE IS THE LINK. Some ideas never die, they just evolve, king. Like you.
Please, I'm very cold.
Unfortunately there are no pictures of the rear exterior of this house, so this is where we will have to conclude for today. That being said, these houses and their antecedents are developing a design language all their own that will, in time, be as culturally rich to us as the houses of yore. The problem is they are less visually interesting. They are houses made to scroll in and scroll right by. Expect to see more of them here, but only if they have something, anything to say.
If you like this post and want more like it, support McMansion Hell on Patreon for as little as $1/month for access to great bonus content including a discord server, extra posts, and livestreams. (Donât worry! This doesnât adjust for inflation! Nowâs the perfect time to join!) By the way: new subscribers can buy a year of McMansion Hell for just $12!
Not into recurring payments? Try the tip jar! (I would seriously appreciate any and all tips because I am in the process of moving house!)
areudirty
dededos
baratheonpilled
dailybridgerton
janainaart
prguitarman
