4 ms·

what about all the crypto scams? all the state media sponsored propaganda?

meowster 2y ago
Did they ever remove that 1M+ subscriber chanel that pretends to be SpaceX and show launches only to pivot to a deepfake of Elon promoting a cryptoscam? YouTube allowing that scam channel to get so large is dumbfounding.
- nullc 2y ago
  There is an ongoing lawsuit regarding the Steve Wozniak version of those videos, by the real Steve Wozniak and a number of the victims. Youtube's efforts to get it dismissed completely on s230 grounds has failed, apparently on the basis that youtube awarded verification badges to the scammers and also running targeted advertising on behalf of the scammers.
  - Nasrudith 2y ago
    I worry about what that would do to future viability of verification and chilling future verification efforts. Lawyers once again being the reason why we cannot have nice things due to not doing so being the safe option. Now, a flawed verification process isn't a good thing but flaws would necessarily come with widespread access even if there was some sort of government support infrastructure for it. (The DMV doesn't prevent the existence of fake IDs.) A situation where verification is maximally reliable above all else would be "literally give it to just heads of state", would open up the exact sort of harms that lead to verification being a thing in the first place. Like being able to Spartacus Steve Wozniak.
    - ThoriumClock 2y ago
      Perhaps that's going down a slippery slope, and there's some happy medium between YouTube's current verification process and a maximally reliable verification process. Indeed, the very first paragraphs of Wozniak's lawsuit seek to set out a contrast between an appropriate response and an inappropriate response: On July 15, 2020, Twitter suffered a massive hack that hit 130 Twitter accounts of celebrities and public officials, including Joe Biden, Barack Obama, Elon Musk, and many others. The attackers sent tweets from those accounts offering to send $2,000 for every $1,000 sent to an anonymous bitcoin address. That same day, Twitter acted swiftly and decisively to shut down these accounts and to protect its users from the scam, issuing the above apology. In stark contrast, for years, Defendant YOUTUBE’s lax security practices have led to a steady stream of security breaches in which countless popular YouTube channels have been hijacked and taken over by criminals who use the channels to perpetuate a scam that has defrauded, and continues to defraud, YOUTUBE users of millions of dollars. Several of the Plaintiff's causes of action, such as negligence, would allow for a far more limited ruling than to be applicable any time a company accidentally miscommunicates about the identity of another person. For example, it could be negligent design that after 6 years of these scams, YouTube still allows hacked verified accounts to change their name to "MicroStrategy" or "SpaceX", without flagging or limiting usage of those identities due to the fact those names are used daily to run scam videos. Good regulation often seeks a delicate balance, striving for a happy medium that satisfies no one entirely but works well enough for everyone.
    - nullc 2y ago
      I also have "this is why we can't have nice things" response, but flipped around: the grossly negligent, even outright abusive behavior by a few companies are going to cost us S230. This sucks all the more because google's legal power means they hardly need S230 while smaller sites such as HN or Wikipedia, for example, are much more in need of it. (and at the same time these other sites are much better at not being embarrassingly bad at moderation) The best situation I think is where there is robust protection in the law to prevent the huge waste of litigation, while at the same time sites work diligently to remove unambiguously illegitimate material and at least behave compassionately about whatever slips through the cracks. But that's not what youtube has done, they've hardly even taken basic measures. E.g. take down obvious scam content when mass reported, block reuploads of bit-identical copies (Or matches with the same perceptual hashes they use for copyright takedowns as part of their covenant not to sue agreements with studios and record labels which they use to rob artists), block varrious low false positive scam identifying keywords associated with the scam, remove the validation flag when an account changes names. etc. Having missed things like this the first time around might be excusable, but to refuse to take measures like this after years of abuse while throwing up a hard S230 wall to seemingly justify continuing to distribute the scams. It's outrageous, it's offensive, and it will ultimately deprive us all of S230 protection. > all else would be "literally give it to just heads of state", Twitter had some "government official" badge which they happily gave to a bunch of crypto grifters after they just invented a fake country.

Did they ever remove that 1M+ subscriber chanel that pretends to be SpaceX and show launches only to pivot to a deepfake of Elon promoting a cryptoscam? YouTube allowing that scam channel to get so large is dumbfounding.

nullc 2y ago
There is an ongoing lawsuit regarding the Steve Wozniak version of those videos, by the real Steve Wozniak and a number of the victims. Youtube's efforts to get it dismissed completely on s230 grounds has failed, apparently on the basis that youtube awarded verification badges to the scammers and also running targeted advertising on behalf of the scammers.
- Nasrudith 2y ago
  I worry about what that would do to future viability of verification and chilling future verification efforts. Lawyers once again being the reason why we cannot have nice things due to not doing so being the safe option. Now, a flawed verification process isn't a good thing but flaws would necessarily come with widespread access even if there was some sort of government support infrastructure for it. (The DMV doesn't prevent the existence of fake IDs.) A situation where verification is maximally reliable above all else would be "literally give it to just heads of state", would open up the exact sort of harms that lead to verification being a thing in the first place. Like being able to Spartacus Steve Wozniak.
  - ThoriumClock 2y ago
    Perhaps that's going down a slippery slope, and there's some happy medium between YouTube's current verification process and a maximally reliable verification process. Indeed, the very first paragraphs of Wozniak's lawsuit seek to set out a contrast between an appropriate response and an inappropriate response: On July 15, 2020, Twitter suffered a massive hack that hit 130 Twitter accounts of celebrities and public officials, including Joe Biden, Barack Obama, Elon Musk, and many others. The attackers sent tweets from those accounts offering to send $2,000 for every $1,000 sent to an anonymous bitcoin address. That same day, Twitter acted swiftly and decisively to shut down these accounts and to protect its users from the scam, issuing the above apology. In stark contrast, for years, Defendant YOUTUBE’s lax security practices have led to a steady stream of security breaches in which countless popular YouTube channels have been hijacked and taken over by criminals who use the channels to perpetuate a scam that has defrauded, and continues to defraud, YOUTUBE users of millions of dollars. Several of the Plaintiff's causes of action, such as negligence, would allow for a far more limited ruling than to be applicable any time a company accidentally miscommunicates about the identity of another person. For example, it could be negligent design that after 6 years of these scams, YouTube still allows hacked verified accounts to change their name to "MicroStrategy" or "SpaceX", without flagging or limiting usage of those identities due to the fact those names are used daily to run scam videos. Good regulation often seeks a delicate balance, striving for a happy medium that satisfies no one entirely but works well enough for everyone.
  - nullc 2y ago
    I also have "this is why we can't have nice things" response, but flipped around: the grossly negligent, even outright abusive behavior by a few companies are going to cost us S230. This sucks all the more because google's legal power means they hardly need S230 while smaller sites such as HN or Wikipedia, for example, are much more in need of it. (and at the same time these other sites are much better at not being embarrassingly bad at moderation) The best situation I think is where there is robust protection in the law to prevent the huge waste of litigation, while at the same time sites work diligently to remove unambiguously illegitimate material and at least behave compassionately about whatever slips through the cracks. But that's not what youtube has done, they've hardly even taken basic measures. E.g. take down obvious scam content when mass reported, block reuploads of bit-identical copies (Or matches with the same perceptual hashes they use for copyright takedowns as part of their covenant not to sue agreements with studios and record labels which they use to rob artists), block varrious low false positive scam identifying keywords associated with the scam, remove the validation flag when an account changes names. etc. Having missed things like this the first time around might be excusable, but to refuse to take measures like this after years of abuse while throwing up a hard S230 wall to seemingly justify continuing to distribute the scams. It's outrageous, it's offensive, and it will ultimately deprive us all of S230 protection. > all else would be "literally give it to just heads of state", Twitter had some "government official" badge which they happily gave to a bunch of crypto grifters after they just invented a fake country.