doublekill

What happened during start of pandemic is WHO and big tech meeting up to discuss the threat of an "infodemic". Companies such as Google, Facebook, and Twitter, agreed to suppress amateur takes on COVID. Mentioning COVID would vastly decrease your reach if not a verified public health source. It would get you demonitized, to not incentivise low-quality clickbait COVID videos. Of course, these algorithms are way more aggressive and produce more false positives (ads removed for just mentioning COVID without giving false cures, like happened here).

It has nothing to do with white or black and race. Just the nitwits at WHO and big business mismanaging crowd intelligence during a pandemic. WHO said masks do not work. Youtube banned users for mentioning vitamin D defiency and COVID. But they meant it well...

An Israeli had build a tool to detect anti-semitism in videos and comments. Not being able to sell to YouTube, they pivoted to targeting advertisers ("make sure your advertisements do not appear next to racist content"). Part of this was a PR campaign, where they would mail journalists "scoops" with videos with lots of views, anti-semetic content, and no moderation of YouTube.

With every such news article, YouTube stepped up its game. It culminated with PewdiePie making an "anti-semetic" joke, and this resulted in the Adpocalypse.

So I'd say it was more commercial interests than activism interests, but at times these align.

She tried to bypass internal review. She knew that internal review was part of the process. She auhored a paper under the Google banner, not making note of her colleagues' work which combat the bias she was decrying. Then she submitted it for internal review the day of the submission deadline. After internal review was done and suggestions to improve were made (valuable time from your peers!) the paper was already accepted by external reviewers. She was then miffed having to update or retract the paper.

Jeff Dean is an absolute saint and ally for the cause. For sure he has better people skills than Gebru. Jeff was over his head with the Twitter mob, which would mess with anyone, no matter how famous.

Google fired Gebru not for retaliation, or for critical science, or for being Black, or for advocating diversity, or for Dean having bad people skills. She was fired for being insufferable.

Only astroturfing Ive seen is from activists claiming racism or Jeff Dean autism / white privilege.

This is just an experiment to see how long it takes for a random comment on a submission currently ranking 533 will take if it contains the word "chinese bioweapon". Feel free to ignore if you happen to be in this thread on coincidence and sorry if detracting from real conversation.

https://thebulletin.org/2020/06/did-the-sars-cov-2-virus-ari...

A very troubling idea if true.

The journalist won't get to talk to the neurodiverse person. Usually someone lacking a thick skin or easy to take offense, goes to HR to complain (anonymously if need be) about offensive, awkward, sexist, racist, far-right, ... speech. The neurodiverse person will (rightly) get a warning, an internal investigation finds no systemic harm done, and that's that, you won't even hear about it, as it juridically not smart to be transparent about such cases.

Then the offended person changes jobs, complains on Twitter about sexist humor driving her out of her previous job, a journalist searches for leads on their next story, and suddenly you are in the news with "multiple people complained about sexist and racist humor on the work floor, but nothing was done about it, and no complaint or investigation resulted in any action. We asked the company for a response and they replied that these internal investigations turned up nothing and that they don't accept discrimination of any kind".

Nearly every company has incidents of sex between co-workers, or people taking illegal drugs on a company get together. Depending on who you get to talk to that's "it seemed like every other week I saw a used condom in the stair ways" or "management regurarily used cocaine during company parties". You won't talk to anyone relativating it, and company PR is shy to even admit that stuff.

My first European colleague was a male-to-female person. She had worked at a major company for years, got a bad performance report, and attributed it to her transition. It went to court and the company settled, instead of defending not promoting her (with or without regard of her transition). Very good programmer. She worked there for 1 year more, doing only things that the company laywers agreed to.

In America I see similar things. I think it is seductive to be swept up with identity politics, and suddenly that promotion that went to a white person, feels different, feels racist. Or you employ a neurodiverse low-social skill person who makes an edgy unfunny joke, because he is nervous. That's "racist humor on the work floor" when going to a lawyer or journalist. It is a dangerous world out there. Current climate not helping.

What I learned from diversity training is that you need to accept the fact that society is systematically racist, and as a white person you benefit from this. So all whites are profiting from racism already. They have to come to terms with this, acknowledge the role that their ancestors played in furthering or creating racial inequality, and do everything in their privileged power to right this wrong: hire more black people. Promote more black people. Mentor more black people. Pay reparations to black people. Be a vocal supporter to black people. And realize and appreciate that you are different, and will never have to share what black people go through every day. What more can you do?

I just did. The research is divided into two: those that look at clinical outcomes (such as cardiac arrest) do not find that vitamins help. Those that look at quality of life indicators, recovery, infection severity, and muscle strength do find significant improvements of supplementation.

And they find no harm of normal use, but warn against dangerous toxicity effects of overdosing.

90 percent of the restaurant is eating, but just 20 percent of those are still hungry :). They can stop eating and return next week when they are starving.

For me it is proof enough that my wise mother told me take my vitamins, but more scientific: you have a vector of essential vitamins and minerals. There are thresholds for deficiency. There are optimal levels. Aim for optimal levels, and do not overdo it. Supplements can help with that, and easier and cheaper than a diet which includes all necessities. Low levels are suboptimal for you. Makes sense?

Deficiencies disparately impact minorities and poor people. And when I was a student even I could not afford that Bell pepper. 1 in 5 or 1 in 10 are real people.

If you take the prescribed one pill a day, how are you going to form overdose? Talking about excess I hear the medical professional disdain for vitamins in that. Understandably, because they get those people who take extreme doses of over the counter stuff in their visiting hours. But that's no risk to use as an argument against vitamin supplementation for regular use. Multivitamins would be banned if regular use caused overdose.

So what about:

> The Centers for Disease Control and Prevention found that nearly 10% of all Americans have a nutritional deficiency.

And what about very low, but non-deficient levels?

It very much depends how the study is designed: in this case they do not measure quality of life or disease progression, but clinical outcome. That is what OP was alluding to: if you are going to only look at clinical outcome and completely tar "effectiveness" with that, then that is a very one-sided view of health and well-being.

Osteoporose patients are at increased risk of vitamin K deficiency. Just because multivitamins are not effective at taking us to Mars, are you really going to recommend patients to spend their vitamin money on fresh fruits?

They only measure clinical outcomes such as cancer or heart attacks. "Multivitamins do not cure cancer" would be a better heading. Or even "30% of Americans taking multivitamins report enhanced quality of life".

Terrible study if that is the conclusion. Yes, multivitamins don't help with cancer or heart attacks. Yes, it is better to eat a healthy diet.

But vitamins do physically help when you are deficient. Your energy levels and immune system improves.

And "all in your head" is nothing to scoff at when the Placebo effect has been measured to help shorten disease by 7%.

Having someone else do your prison time was fairly common. Now with more cooperation between organizations and improved biometrics it has dropped.

Different people would be arrested, show up to court, or enter the prison. Impossible for organizations downstream to detect or assume a switch had happened.

Vitamin C is listed as fake news on WHO. Fact checker sites listed vitamin D as false, then switched to correct after research. Doctors, journalists, and health experts in Brazil and the Netherlands chided social media for not removing "fake" info on vitamin supplements.

I'll check some CDC sources later to contextualize these claims to the US.

Experts and health authorities were adament to tell us that vitamins have zero effect against COVID. Not: we don't know and taking a Vitamin C can't hurt, unless you count an upset stomach. But: stop sharing fake health information, this is an infodemic! Just wash your hands.

So to add to the infodemic: selenium and iodine deficiency also increases severity. Take some iodized salt and Brazil nuts now, or wait 5 months for the authorities to understand that absence of evidence is not evidence of absence. And no matter what Youtube bans you for going against the WHO: tumeric is an efficient antiviral.

We can use 90% of our brains at the same time, but the medical term for that is a seizure and it seems to involve a lot of involuntary movements.

Agreed. Also, nobody is, or should be, using deep neural networks, for legislation and law enforcement. Explainability should be a core design decision when making an algorithm, and not slapped on top of an inherently black box algorithm. Black boxes and even their explanations are used to launder bias and unfairness. And most of these tricks are not even explanations that can be trusted. "Oh look, the cat's head is highlighted, so that's why this picture was classified as a cat!" no insight, no justification, just hoping the network learned some higher level features like humans do, but oh no, when we flip the picture it is suddenly a dog, and when we photoshop the background to be snow, now it is suddenly a polar cat or a pinguin.

Let deep learning do what it is good at, without explaining their performance and errors to anyone: invading your privacy on social networks, helping hedge funds make more money by analyzing Elon Musks tweets, and building military surveillance.

Leave the justifications and explanations to inherently white box models (they are nearly as good in performance as black box now, at least for structured data), and hold off on firing radiologists for a few decades, even though your train set performance is overfitted to be on par with "human-level".

Somehow, somewhere, the deep learning revolution started to drink its own kool-aid and became alergic to critique or solid verifiable computer science. Explainable deep learning does not exist, since half of the time the engineer that build the system can't even explain why it works in the first place. "Strong inspectable feature engineering is hard and time-consuming, so here we shook a box of legos a million times, burned six holes in the ozon layer, and out comes a deep net optimized with gradient descent". End-to-end learning is supposed to be really end-to-end, including the explanation.

Move away from one-size-fits all, but have the UX adapt to user behavior. Personalized UX is pretty close to perfect.

No, there is a 2009 study which compares N95 mask efficiency with surgical masks, because they predict that during a pandemic N95 masks will be in shorter supply. (1)

They don't find significantly more efficiency for N95 masks, but other studies do find N95 masks slightly better. (2)

(1) https://jamanetwork.com/journals/jama/fullarticle/184819

(2) https://onlinelibrary.wiley.com/doi/full/10.1111/j.1750-2659...

Do a search on Google Scholar for "N95" in combination with "virus" or "sars" for more studies.

Really a shame that authorities, including the Surgeon General, claim that masks are not protective, despite medical science to the contrary.

Deep learning is mostly focused on image recognition, and classification on tabular data is way more common for smaller businesses.

Your activity online triggers 10s of AI models for every hour. AI is just not much of a consumer technology, but it is what the producers increasingly use.

GPT-2 is useful way way beyond text generation, it has very powerful embeddings useful for many other tasks. The text generation and cat picture classification is what pulled it into the mainstream: how you know about it, while not familiar with the particulars.

Stop putting the sole blame on recommendation algorithms for radicalization. Or paint everything with the same brush and start blaming Youtube too for radical thoughts on climate change, veganism, identity politics, etc.

I suspect your brother is not even a Nazi anyway, just has differing views on immigrants and black crime statistics. Let's hope some other Youtuber does not get recommended a video on how to punch a Nazi, as by your reasoning, people could not help themselves.

The newspapers of the time also dubbed the Spanish flu: just a familiar grip we've seen time and time again.

China desperately wants to save face, at the cost of lives. They rejected help from the U.S. CDC as to not appear weak. They already thought they had it under control, before locking down over 50 million people and damaging their economy.

People at risk for the flu can get a flu shot. This disease kills the healthy and average aged. And we all have family members at increased risk.

Even a 2% mortality rate is absolutely horrific. Would you play Russian roulette with those chances? If 60% percent of the world gets infected we are talking many many millions of dead.

How are you going to track contacts, when someone working in a brothel gets it, or someone preparing food in a McDonald's? What if someone takes the virus back to India or Africa?

With 20% needing intensive care, even modern European countries will suffer badly. There simply are not enough beds available, even in the bigger cities.

Mortality is a red herring when you do not count the people still hospitalized after months! Look at the tiny amount of recovered people.

Despite their titanic and gruesome methods to contain it, China has failed. Imagine such an effort in Europe, where people gladly will walk away from quarantine and can not be cowered into compliance or snitching on their coughing neighbors.

We simply do not know enough about morbesity and mortality to say we overreact. Better safe than sorry when uncertainty abound.

Did he not donate to a legal political cause, without publicizing it?

I doubt funders care about your opinions on gay marriage, but they do care a lot about the stink the cancel culture raised around him.

The poison seems to be applied externally.

Your facial features don't change after you do a crime (except maybe for some new prison tats). They don't need to change to keep criminal classification from head shots.

Age, race, gender, income, attractiveness, testosterone levels, gang membership, attention to grooming, addiction, etc. are all predictive of crime, and part of the face picture you use.

Sure, there is some bias in how you label a criminal (convicted of a crime, self-report, etc.), but prediction is possible (without exploiting leakage like smiles or lighting).

Just from a hot-or-not rating I can make an educated guess of your conviction rate and sentencing length.

Have you verified that this is impossible or was gathering evidence for your strong claim too obvious for you?

It is possible to do this better than random guessing!

Did not downvote, I hate all views equally.

Perhaps it is the enormous cognitive dissonance. Western companies also take military funding and their surveillance systems were always better and more invasive.

The US is using statutes for national security interests to ban Chinese companies for contributing to alledged human rights abuses. Meanwhile we have documented human rights abuses in the West, and a US that does not shy away to cover a trade war and skewed commercial motives with anti-China propaganda.

US companies and agencies surveil outside the US. Capitalism-commercialization gave our rights to the likes of Zuckerberg and Amazon, who will turn your door bell into a police camera.

The Godwin was also a tad unnecessary. Perhaps you are a bit caught up in the heat of the moment. The propaganda machine is definately turning up the heat, so I do not blame you. Try to pose your argument without claiming the moral highground of the "free world". It is never fortunate that an entire country and its people struggles.

Accusing Ng and Altman of helping build a fascist surveillance state is a nasty accusation, which I feel you are not allowed to make.

Remember when the US had concentration camps for Japanese-Americans? Countries can change even without pressure or backlash (the US did not face any). Can you think of a modern group of people in the US being secluded from participation, facing state racism, and being locked away in ghettos, with no hope of social movement? Would China be justified in boycotting or sanctioning US companies for contributing to ICE, Guantanamo bay, or police violence, or Flynt's water, or Trump's racism?

Freedom of religion gave the US scientology and TV pastors and schools preaching creationism. It gave Europe extremist Salafism and terrorist attacks and people using the Bible to go after immigrants. You have to see China in that context: they see religion as a dangerous memetic virus to indoctronate youthful people into believing something not of their own choice. Countries have a right to make their own laws and their own tactics for improvement and combating extremism.

Edit: you may also receive downvotes from people who hate to hear negative stuff on China, or are paid 50 cents to do so.

I do not think it is the SEO, but Google adapting to its less-technical userbase (people who search for "facebook" and click the first result).

It makes sense for this target audience to give broad results. They seem to be relying more on neural language models to accomplish this.

Google also surfaces more authoritative results than Bing, Google thus favors bigger commercial sites.

Google is suprisingly good on vague searches (for instance, describing a singer's appearance and song themes will return the name), but getting worse on "conspiracy"-type searches (you will not find many amateur sites discussing the high-class escort past of the First Lady, because these lack authority, all results focus instead on the massive fine for posting allegations).

Up until a few years back, crypto was left completely behind. Still possible to beat the quants and the suits, iff you know where to look.

Because it is not a social science paper or a government policy (smoking causes cancer, so we should forbid advertising smoking to teenagers). You do not combat shoplifting by knowing the causes of shoplifting (being unethical causes shoplifting, now what?).

Nobody infringes on your rights, if a system pings you as suspicious and so a security guard is alerted and watches you more closely. There are no false positive consequences here: either they find that can of Red Bull on your person without a paid receipt, or they don't. You won't get banned from the store, because the system inadvertently deemed you suspicious.

A whitebox statistical model still can provide the factors that contributed to its prediction, without resorting to causal inference.

I do not think a certain race, gender, age, income causes shoplifting, but they sure are correlated, and effective at finding shoplifters or tax fraudsters. I deem the act of shoplifting more unethical than the act of watching you on a camera feed. This already happens anyway, these systems just manage attention better.

Sucks to be poor, but that can never be an acceptible excuse for criminal behavior (as long as you can feed yourself and your family through legal ways, shoplifting is hardly a desperate necessity for first world inhabitants, but damages from shoplifting do bankrupt hard working store owners).

If the discrimination (not racism!) causes a bias towards a certain race, neglecting other races who shoplift, and focusing on innocents, then the accuracy will suffer. These systems are not deemed problematic because they are ineffective, they are problematic, because they do a "good" job.