8 ms·

That makes sense. The problem is that at the end of the day the decision will have to be made either by algorithms or by people. Algorithms are less nuanced, ha

That makes sense. The problem is that at the end of the day the decision will have to be made either by algorithms or by people. Algorithms are less nuanced, hard to write, and will often fail, people are expensive and biased. With the amount of videos youtube has to process, and without an almost human-level AI, I don't think there's a perfect solution. Algorithms flagging videos for review, and then humans making decisions based on some basic set of principles seems right. Also, to let advertisers decide where to spend their money there's targeting tools...

hkmurakami 9y ago
Aren't algorithms also biased in theory since they are created by people?
- lc5G 9y ago
  While algorithms can be biased intentionally or unintentionally, it is not true that ALL algorithms HAVE to be biased because they are created by people. In machine learning it might be easier to unintentionally create biased algorithms. On the other hand there are domains where I would trust algorithms to be less biased than people.
- drchickensalad 9y ago
  The fact this is often glossed over causes many problems in society. https://99percentinvisible.org/episode/the-age-of-the-algorithm/ https://99percentinvisible.org/episode/the-age-of-the-algori... makes some interesting and good arguments imo.
bufferoverflow 9y ago
The solution is obvious - tag/categorize the videos, according to very strict rules, let the advertisers decide which tags to serve. Allow the publishers to dispute the tags. For instance, "nudity" can be defined as "visible genitals, female nipples" in the US. "Controversial" tag should not even exist, it makes no sense, any non-trivial subject is controversial. And honestly, YouTube can benefit from a more precise tagging/category system, it's currently extremely crude.
- aestetix 9y ago
  I think a better solution is to have some core principles that Youtube's censorship team abides by. If the video in question doesn't violate all the principles (or X number of them), then it remains up. Of course that is difficult and actually involves discussions, which Youtube doesn't seem interested in having.
  - pr0zac 9y ago
    There is zero censorship involved here. This isn't a leak about deciding what gets taken down, its about what is able to be monetized. The leak in fact specifically mentions this choice was used to avoid censorship by removing video.
- freeflight 9y ago
  Afaik they already thought of that solution because YouTube videos support tagging/categories but it's the video uploader who sets those tags. The problem is that once you introduce such a system, and leave all the control over the tags to the uploader, people start gaming it for better search rankings/more views. So you are back to the same old problem of "who checks if the content actually matches the tags".
  - awiesenhofer 9y ago
    I think in bufferoverflows solution the tags come from algorithms and user reports not the publisher.
- amarkov 9y ago
  Strict rules for tagging would just run into the same problems as strict rules for demonetization. LGBT groups won't be any happier to see their content publicly branded "sexually explicit" than just demonetized.
  - bufferoverflow 9y ago
    By strict rules I also mean objective and unbiased. Penis/vagina -> nudity (even if it's medical or educational). "Fuck" -> cursing (even if it's about the etymology of the word). Advocating murder of group X -> ban (even if you're quoting a religious text). If these rules are strict and clear, we have a chance of creating a legal-like system. Otherwise it's subjective chaos with everyone pissed off.
notyourday 9y ago
> Algorithms are less nuanced, hard to write, and will often fail, people are expensive and biased. Algorithms have a fantastic property of repeatability which means that it is (a) possible to adjust them over time (b) possible to identify what exactly happened (c) blind to the context and inputs outside of the algorithm That's exactly what is needed.
- majewsky 9y ago
  > Algorithms [allow to] identify what exactly happened "This video was demonetized because at the 1:15 mark, there is a patch of light dirt that our neural network classified as naked skin with a 70% confidence." Oh, you meant you were going to do image recognition without any sort of machine learning?
  - notyourday 9y ago
    Your alternative is the idiot on his last day at Twitter who in his post Trump account blocking interview basically admitted that it was the totality of things that happened that day to him and it being his last day and that Trumps account popped up in front of him which caused him to make the arbitrary decision to block it. Please remove humans from making decisions that need to be uniform. So yes, algorithm it is.
    - cnewey 9y ago
      You're implicitly making the assumption that an algorithm is unbiased and less likely to be fallible - and unlikely to be susceptible to bias or manipulation (at least, compared to a human). There is a considerable amount of academic and popular literature showing that these assumptions are false, at least at a technical level. The problem of vetting YouTube videos isn't necessarily something that can be solved by an algorithm, at the end of the day. Lots of the simple cases, sure - but such an algorithm won't work on even the most moderately challenging content. After all, moderating complex content such as video doesn't just require a well-trained neural network to recognise objects portrayed in images and sounds in audio... deciding whether a video is appropriate or not for monetisation is something that requires an explicit understanding of the social and cultural context which the video is going to be viewed in - which is both highly subjective and difficult to express in an algorithm. NB; - Neural networks (and other AI techniques) can be tricked and manipulated (see [0,1]; examples of adversarial images) - Machine learning algorithms can induce or confirm bias through inadequate training data and poor assumptions. (see [2]; a book about bias in machine learning algorithms) [0]: https://arxiv.org/abs/1510.05328 https://arxiv.org/abs/1510.05328 [1]: http://www.evolvingai.org/files/DNNsEasilyFooled_cvpr15.pdf http://www.evolvingai.org/files/DNNsEasilyFooled_cvpr15.pdf [2]: https://weaponsofmathdestructionbook.com https://weaponsofmathdestructionbook.com

hkmurakami 9y ago

Aren't algorithms also biased in theory since they are created by people?

lc5G 9y ago
While algorithms can be biased intentionally or unintentionally, it is not true that ALL algorithms HAVE to be biased because they are created by people. In machine learning it might be easier to unintentionally create biased algorithms. On the other hand there are domains where I would trust algorithms to be less biased than people.
drchickensalad 9y ago
The fact this is often glossed over causes many problems in society. https://99percentinvisible.org/episode/the-age-of-the-algorithm/ https://99percentinvisible.org/episode/the-age-of-the-algori... makes some interesting and good arguments imo.

bufferoverflow 9y ago

The solution is obvious - tag/categorize the videos, according to very strict rules, let the advertisers decide which tags to serve. Allow the publishers to dispute the tags. For instance, "nudity" can be defined as "visible genitals, female nipples" in the US. "Controversial" tag should not even exist, it makes no sense, any non-trivial subject is controversial. And honestly, YouTube can benefit from a more precise tagging/category system, it's currently extremely crude.

aestetix 9y ago
I think a better solution is to have some core principles that Youtube's censorship team abides by. If the video in question doesn't violate all the principles (or X number of them), then it remains up. Of course that is difficult and actually involves discussions, which Youtube doesn't seem interested in having.
- pr0zac 9y ago
  There is zero censorship involved here. This isn't a leak about deciding what gets taken down, its about what is able to be monetized. The leak in fact specifically mentions this choice was used to avoid censorship by removing video.
freeflight 9y ago
Afaik they already thought of that solution because YouTube videos support tagging/categories but it's the video uploader who sets those tags. The problem is that once you introduce such a system, and leave all the control over the tags to the uploader, people start gaming it for better search rankings/more views. So you are back to the same old problem of "who checks if the content actually matches the tags".
- awiesenhofer 9y ago
  I think in bufferoverflows solution the tags come from algorithms and user reports not the publisher.
amarkov 9y ago
Strict rules for tagging would just run into the same problems as strict rules for demonetization. LGBT groups won't be any happier to see their content publicly branded "sexually explicit" than just demonetized.
- bufferoverflow 9y ago
  By strict rules I also mean objective and unbiased. Penis/vagina -> nudity (even if it's medical or educational). "Fuck" -> cursing (even if it's about the etymology of the word). Advocating murder of group X -> ban (even if you're quoting a religious text). If these rules are strict and clear, we have a chance of creating a legal-like system. Otherwise it's subjective chaos with everyone pissed off.

notyourday 9y ago

> Algorithms are less nuanced, hard to write, and will often fail, people are expensive and biased. Algorithms have a fantastic property of repeatability which means that it is (a) possible to adjust them over time (b) possible to identify what exactly happened (c) blind to the context and inputs outside of the algorithm That's exactly what is needed.

majewsky 9y ago
> Algorithms [allow to] identify what exactly happened "This video was demonetized because at the 1:15 mark, there is a patch of light dirt that our neural network classified as naked skin with a 70% confidence." Oh, you meant you were going to do image recognition without any sort of machine learning?
- notyourday 9y ago
  Your alternative is the idiot on his last day at Twitter who in his post Trump account blocking interview basically admitted that it was the totality of things that happened that day to him and it being his last day and that Trumps account popped up in front of him which caused him to make the arbitrary decision to block it. Please remove humans from making decisions that need to be uniform. So yes, algorithm it is.
  - cnewey 9y ago
    You're implicitly making the assumption that an algorithm is unbiased and less likely to be fallible - and unlikely to be susceptible to bias or manipulation (at least, compared to a human). There is a considerable amount of academic and popular literature showing that these assumptions are false, at least at a technical level. The problem of vetting YouTube videos isn't necessarily something that can be solved by an algorithm, at the end of the day. Lots of the simple cases, sure - but such an algorithm won't work on even the most moderately challenging content. After all, moderating complex content such as video doesn't just require a well-trained neural network to recognise objects portrayed in images and sounds in audio... deciding whether a video is appropriate or not for monetisation is something that requires an explicit understanding of the social and cultural context which the video is going to be viewed in - which is both highly subjective and difficult to express in an algorithm. NB; - Neural networks (and other AI techniques) can be tricked and manipulated (see [0,1]; examples of adversarial images) - Machine learning algorithms can induce or confirm bias through inadequate training data and poor assumptions. (see [2]; a book about bias in machine learning algorithms) [0]: https://arxiv.org/abs/1510.05328 https://arxiv.org/abs/1510.05328 [1]: http://www.evolvingai.org/files/DNNsEasilyFooled_cvpr15.pdf http://www.evolvingai.org/files/DNNsEasilyFooled_cvpr15.pdf [2]: https://weaponsofmathdestructionbook.com https://weaponsofmathdestructionbook.com