YouTube Knows How to Stop Serving Toxic Videos

>People who watched video A and also watched video B and were very likely to comment and interact with it, therefore people watching video A should be presented with video B.

I don't think this is how YouTube's recommendation algorithm has worked in a long time. I won't argue that the current recommendation algorithm is somehow nefarious in nature, but I'm willing to bet there are factors that are weighed a lot heavier than the fact "I watched A so I should watch B". Your recommendations hardly change based on what video you are watching at the current moment.

My personal theory is that YouTube overemphasizes watch time to a point where any long video gets suggested as long as it has a high completion time. You could be "batshit liberal" and still be stuck in the "alt-right recommendation loop" simply because YouTube thinks you will finish those videos regardless of your political orientation.

Why does YouTube think I want to watch "Student Shuts Down Social Justice Warrior College Professor"?

I'm conservative(ish) in a fiscal sense, but don't read conservative news (because it sucks), don't search for conservative topics (nor really any political topics), and am not interested in any alt-right or political conflict type videos. Yet a smattering of this kind of thing often winds up in sidebar suggestions.

I can't help but feel it's a bit of a fishing expedition to determine political alignment. Probably paranoid though and it's more like conflict gets eyeballs to serve ads to. Which is a statement on human nature more than Google. Who knows, but it is annoying.

> Too frequently, the people writing about this and the politicians talking about it really just want to be the ones to set the line on what is "toxic" and "offensive" and have a very strong desire to influence what Google and these other companies should censor.

I've definitely seen this pattern from people. It's rarely overt, but I do see demands that others "punch down" or that "toxic" things be treated a certain way. This is always coupled with an implicit assumption that there's some clearly understood definition of what those mean that just happens to be exactly what the person using those idioms would offer.

Like you, I think this pattern suggests people trying to grab for the privilege of writing definitions.

Private company: Not censorship.

And, yeah, it's tough to specifically define what's "toxic" and/or "offensive." But YouTube's algorithms are unforgivably dumb about it.

You are missing the point of the piece. The author is saying YouTube has the ability to not show content that is flat-out wrong and dangerous. The author is saying the same thing you are. The algorithms can be easily adjusted - but they wont be, because controversy = time on Youtube. They absolutely do not have to recommend content that would radicalize a man enough to take a rifle into a pizza parlor to free alleged child sex slaves.

This is not about setting the line about what is "offensive." this about taking responsibility for providing a huge platform for dangerous ideas to spread.

No Im not advocating for thought police but I really do think most would agree that platforms have to start taking responsibility for promoting QAnon style conspiracies and white-supremacist ideas.

Good point. It does seem like these people just want their "toxic" videos visible and the other side's "toxic" videos censored. At the end of the day, it's all about control and propaganda.

How about hmmdaily and the people who are against "toxic" video just put out interesting videos? Why not compete instead of trying to censor everything?

Also, I've never heard of hmmdaily. If anyone knows what it is and what their agenda is, I'd love to learn more about it.

Do you have any basis for claiming that Google Video Search only shows whitelisted videos? I thought only Google News was limited that way.

I did a few quick searchs on the video tab, and pulled up a lot of videos from YouTube and at least one from Know Your Meme and another from Giphy. None of them seem like obvious whitelist candidates.

Ironic that the author is accusing YouTube of profiting off of conspiracy theories... by promoting his own conspiracy theory.

I think it is a matter of target metric. You can optimize just for click-through/watch time (short term reward), but you can also target a composite of that and quality/trustworthiness (long term reward). YouTube seems to mostly optimize the former, while search seems to compromise between both.

Consider the possibility that turning Youtube into Google News is the goal of the people who talk like this.

The internet democratized publishing. That made a lot of people unhappy - not just because of nonsense like Q, but because it let people say they weren't too keen on Communism in China, Ben Ali in Tunisia, or Clinton/Macron-style progressivism in the West. China went and un-democratized it, and a lot of people want the same thing to happen here.

It's like how social conservatives wiped out queer communities in places like LiveJournal by targeting NSFW content - they can't come right out and say they liked it better when the only easily accessible sources of information were two local papers and three TV channels, all of which could be depended on to (for example) come out in favor of the war in Iraq, just as the social conservatives couldn't say they wanted to wipe out the fags. So they're targeting conspiracy videos instead.

> but most of it is meaningless noise then we will have less freedom than before

What does this mean? In concrete, practical, down-to-earth sense?

> we cannot be sure the any information we perceive is trustworthy

This point was already made ages ago. Most famously, by ancient Greek skeptics, and by Descates’ cogito. We live in a world where anything may turn out to be untrustworthy. But we somehow cope, and chug along just fine.

"[...] But in the current, digitized world, trivial information is accumulated every second, preserved in all its triteness. Never fading, always accessible. Rumors about petty issues, misinterpretations, slander... All of this junk data preserved in an unfiltered state, growing at an alarming rate. It will only slow down social progress, reduce the rate of evolution. You seem to think that our plan is one of censorship. What we propose to do is not to control content, but to create context. The digital society furthers human flaws and selectively rewards development of convenient half-truths. [...] You exercise your right to 'freedom' and this is the result. All rhetoric to avoid conflict and protect each other from hurt. The untested truths spun by different interests to churn and accumulate in the sandbox of political correctness and value systems. Everyone withdraws into their own small gated community, afraid of a larger forum. They stay inside their little ponds leaking whatever 'truth' suits them into the growing cesspool of society at large. The different cardinal truths neither clash nor mesh. No one is invalidated, but nobody is right. Not even natural selection can take place here. The world is being engulfed in 'truth.' We're trying to stop that from happening. It's our responsibility as rulers. Just as in genetics, unnecessary information and memory must be filtered out to stimulate the evolution of the species. Who else could wade through the sea of garbage you people produce, retrieve valuable truths and even interpret their meaning for later generations? That's what it means to create context."

–"Colonel" AI, Metal Gear Solid 2: Sons of Liberty (2001)

I think the problem lies with the determination of what is and isn't objectionable or dangerous. Who decides? Tech megacorps?

There is plenty of content that we can all agree on being dangerous or worthy of banning, but there is plenty that is not settled, plenty that can blow in different directions with each fresh gust of the political winds.

Should this be handled legally instead, so citizens have recourse?

What happens if the consensus on approved/banned content among the censors is itself dangerous or at odds with reality?

Who is in control of the censorship and what is being censored, what is to be censored, at the biggest, most influential social media venues is an issue that cannot simply be waived away as 'corporate discretion'.

How long until we are all couching our Chinese social credit?

People have never been "in touch" with reality, they're characters in a drama play of their own imagination. Every drama needs villains and heroes.

A few theories. (From the simpliest to the more complicated)

1. The person who wrote this article is very uneducated about this and they're showing their insecurity.

2. They're trying to defer blame of the problem to a pseudo-anonymous group

3. They're trying to avoid putting blame on Mrs. Wojciki due to personal views on identity (identity is a relevant point as that they have a massive bias on what identity and content they want others to find successful)

I'm going to go with #1 because they're pretty preachy.

YouTube isn't an independent "company" in the strict sense of the word (even if it has its own legal entities), it is just a division/product of Google, a subsidiary of Alphabet.

It's impolite to demand personal responsibility from corporations. It was the 'United Fruit Company' that funded dictatorships, and 'Comcast' and 'Disney' that bribed for favorable net-neutrality/copyright laws, and 'Monsanto' and 'Big Pharma' that extort people by abusing and writing in their favor the drug regulatory framework.

Not any CEO or major stockholder with a name and address, goodness no.

This is an argument from personal experience - just because you cant see yourself do it doesnt mean it isnt happening to others.

Clearly since flat earthers continue to exist, your logic is broken.

>Why do you care that they (or any other weird conspiratorial videos) exist at all?

They think their judgement is better than people they disagree with. Therefore someone should exercise paternalistic control over those they disagree with, so those wrong people will stop disagreeing with their obviously-right opinions so darn much.

They also don't foresee the resentment this will generate among the scolded and/or think they can contain that resentment. Which tells me their judgement was actually untrustworthy all along, and regardless of the advisability of censorship they should not be allowed to wield it.

Flat earth is so obviously wrong that I'm safe from it.

I'm not arrogant enough to think that I'm immune to misreporting of facts, or that I and millions of others couldn't be damaged by people arguing in bad faith.

And "arguing in bad faith" covers a lot of the most objectionable content.

It's never them, it's always the "other" people that need to be "protected" from "toxic" content.

Having said that, fake news and conspiracy theories have played a major role in setting the groundwork for the Holocaust, so there's a valid argument to be made there. I wouldn't want the people writing/reading Vox to make the call though.

I don’t care that they exist, but it’s pretty annoying to have them showing up in my recommendations all the time.

To a lesser extent I care about them showing up in other people’s recommendations a lot. Propaganda does work.

YouTube is extremely popular with children and adolescents who generally lack developed critical thinking skills and the context to evaluate the claims in these videos.

Isn't PBS still available, and free?