welcome to a calmer internet

The best Firefox fork in existence. (Gimmick blog, not affiliated with Zen Browser or the Zen Browser team)

[definitely-zen-browser]

Introduction post!

Hey there! I'm definitely-zen-browser (any pronouns) the gimmick blog dedicated to spreading propaganda about the best* browser since sliced bread.

Zen is an open-source fork of Firefox, but with a lot of features and customization options that other browsers don't have. Vertical tabs are one big draw.

Blog run by @scepty-ssb

*my opinion. Also, I don't think there's a sliced bread browser.

@firefox-official @yahooo-official @bingle-official @the-real-google @realsafari

hello other browsers

@bisexual-airforce @gimmick-uncle @tsunderedetector

hello assorted gimmicks

@gimmick-thief you took too long to find my original gimmick so i made another one

[orteil42]

given the current climate this pride especially i feel i must mention that i love my trans friends, i stand with trans people in the fight against transphobic legislation and those who would enforce it, and this blog is not a good place for you to be if you do not vibe with that

[fleshenjoyer]

loss is old enough to vote as of today. i found this out is by making a loss.jpg mii for my tomodachi life island and going to wikipedia to find out its birthday ON ITS EXACT 18th BIRTHDAY. like sure. fuckin whatever i guess

[electronicintrovert]

Learning C has done more to increase my love for Linux than I can adequately express.

It makes poking around the system so much fun, and the man pages make it easy.

And it just feels so much more natural than subprocess or os.system in Python.

Strings would be nice, though.

[mckitterick]

[rip-quizilla]

[banchie]

i've been phasing the phrase 'google it' out of my vocabulary and going back to 'look it up'. fuck you youve lost your generic trademark privileges

[certified-anti-microsoft-posts]

Devastated that unfortunately out of the alternative corporate verbs "Google it" is kinda the best option

[definitely-zen-browser]

Hard disagree, "duckduckgo" already ends with a verb as opposed to the other two which just try to conjugate a noun. Just because "google" happens to fit the most common conjugation pattern in English doesn't make it the best

[utopicwork]

Lets go

[ames-draws]

I'm fed up with "maybe later".

[justalurkr]

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.

[irispurpurea]

This is the paper. It's excellent, highly recommend reading it.

I remember reading about Gebru's firing but I had no idea this was the paper she was fired over.

[01lw]

<faggot> testing something about this godawful website. Someone reblog this please

[01lw]

<very interesting. If you put text in less than/greater than signs then it that text does not show up in the notification> penis

[01lw]

<Very intriguing.> <I have another test.> A

[definitely-zen-browser]

yeah i ran into this a while back, it also works like that on the activity page. i think it's a half-assed attempt at text sanitization to prevent html/script injection

[dragon-in-a-fez]

the thing about being a Tech Person™ is the universe takes it as a challenge. oh you can handle normal problems easily? take some Extremely Not Normal ones. my friend just told me my outgoing voicemail message is in Italian for no apparent reason

[ramun3u]

Yellowish green

ITS JUST YELLOW ITS JUST YELLOW ITS NOT CHARTREUSE ITS JUST YELLOW FUCK ME MAN

[definitely-zen-browser]

☝️🤓 erm ackshually thats slightly dimmer than pure yellow, which would be #ffff00

but nonetheless its exactly 50% red 50% green in terms of hue so its still yellow

[mint-offical]

Maybe be careful with the latest Firefox update.

[pmamtraveller]

Feelings about AI generated art by Veronika Kozlova

[void-milf]

fyi the point of fucking up your data patterns isnt to avoid suspicion. it’s to make EVERYONE suspicious. same logic as the bloc, pals.  protect your comrades, be suspicious. ESPECIALLY if you aren’t doing anything likely to get you arrested.

[void-milf]

the state is less omniscient and significantly more incompetent than you’d think. overextend their resources at every possible opportunity.  make them cry wolf repeatedly. run their data analysis agents fucking ragged.  and strike. attack.

[37womenwieldingknives]

YES i’m a postgrad statistics researcher and i can tell you that the state honestly has NO IDEA what to do with the data it collects, it has an obsession with big data but it’s almost impossible to work with in practice. the traditional statistical approaches that are used can’t be scaled up, the adapted approaches are substantially weakened, and the machine learning approaches have the same problems and often tell them nothing. data scientists are only just coming around to these issues too, most still just push on with it anyway - incompetence is the word. above all this though, like you say, the biggest issue for the state is at the point of data collection. they will NEVER get anything useful if they’re collecting shitty messy data. they will eventually figure out that the real solution is working how to collect accurate and meaningful data, we should make it as difficult as possible for them to do that

[afraidofamericans]

This makes me think that we need WAAAY more apps that generate junk data

[queeranarchism]

Apps that generate junk data? tell me moooore.

[anarchopuppy]

Ooh I know this one!

Ad Nauseum is an adblocker that stores the ads it blocks and continuously generates fake clicks, fucking with analytics and costing the ad companies money

TrackMeNot automatically does randomly generated searches on a variety of search engines to obscure your real searches and fuck with analytics, and you can set it up to work with anything that has a search bar (including facebook, twitter, amazon, youtube, etc)

WhatCampaign replaces analytics parameters in links with the string “FuckOff”. I thought there was a similar extension that used random strings, but I can’t seem to find it

Privacy Possum is a fork of Privacy Badger with a focus on costing tracking companies as much money as possible, and idk if my limited tech knowledge is enough to understand what it does but the description does say it falsifies some data so that’s good enough for me

@afraidofamericans

[mayflower-gal]

Boy it SURE would be a SHAME if this were SPREAD AROUND for everyone TO SEE

[politijohn]

Source

[pingintech]

so this is a pretty common response people have been giving to this and i gotta say myself as someone who worked for annotations for a while on a sub minimum wage as a freelance contractor who solved captchas for robots actively in order to feed this data i too was aware of this but its important to highlight how this is ABSOLUTELY not a thing everyone was aware of and these days captchas are more focused on how you respond to captchas how long it takes you to answer them (and often if you do it too fast youll fail) and it is VERY genuinely a huge wage theft issue not joking because it is fully a job i and many others were paid for meaning it is fully recognized as legitimate payable labor by googles subcontractors and its labor everyone has been asked to do for free for the past 14 years this is an important thing for people to know and something we should all be very upset about

the surveillance state was built by the peoples unpaid labor and works in order to subjugate us directly deprive us of privacy criminalize our existence and replace our jobs

[beguilingcorpse]

[vergess]

A fun bonus fact for you: those No Longer Buyable DVDs?

They're the ONLY surviving NONDAMAGED form of the show. In the late 90s, the masters from which the show is printed were damaged with a red-pink hazy filter.

So. Good luck buying them even if you find them. They're some of the most valuable collector's items in the entire franchise.

Without piracy, there would be NO UNDAMAGED COPIES OF SAILOR MOON AVAILABLE TO ANYONE ANYWHERE, PERIOD.

ETA: Because these undamaged copies are how you colour correct the uncensored ones.

Piracy Is Preservation.