Hacker News Reader

Comment by paulpauper | Hacker News Reader

The among taken in this scam is chump change compared to the YouTube scammers. YouTube is a vastly bigger website than twitter and way slower to respond to accounts begin stolen by scammers. I remember seeing an Ripple giveaway scam that in a single day made 100k with just a single account ,. And fake bill gates one made 40k. the list goes on and on. My guess is the total taken is in the $3-5 million range from youtube alone.

View on HN

And you don't even need to steal an account. When the Playstation 5 launch event was happening I searched for it on Youtube, clicked the top result and it turned out to be a scammer restreaming the real live event with graphics added saying Sony would double your BTC - just send to this address ___.

that may have been an ad. scammers are buying youtube ads too

Ive seen what this previous poster has talked about. They overlay graphics in the actual video stream promoting btc payments. Happened with a Nasa stream right after the first spacex manned mission.

Me too. Youtube even put it in my recommended section.

Definitely not, adverts are blocked here and I had the stream on for a good 30mins.

edit: the layout/graphics were the same as the screenshot in this article linked downthread: https://cointelegraph.com/news/fake-spacex-youtube-channels-...

Archive of the Sony scam site: https://web.archive.org/web/20200612034440/http://ps-btc.inf...

Compare that to the one linked by the compromised Twitter accounts: https://web.archive.org/web/20200715192340/http://cryptoforh...

Those are irrelevant victimless activities. The point of Bitcoin is Eve Online with real money. The only rules are what the protocol says.

It was probably the top result, because the account the hacker was streaming from was a stolen high-profile account.

Good point, I hadn't considered that. Seems it was actually some hours after the real live event[0]. I spot a different channel using the same name[1] that still has a short scam video online from 3 weeks ago.

[0] https://decrypt.co/32102/scam-sony-playstation-5-reveal-even...

[1] https://www.youtube.com/watch?v=N6KECoI7g1U

Youtube served me the Bill Gates bitcoin scam on Monday, 2 days before the Twitter hack, as opening ad for a video from their recommendation algorithm. The ad's site clearly perpetuated the scam for at least a couple days before changing the website to an innocent iframe link to Bill's foundation page.

Alphabet Inc should be held liable.

How do we know the bitcoins was the actual target/purpose? At this point I would be very hesitant to make any sort of judgement like that.

What are these YouTube scams you're referring to? I'm not aware of them.

https://cointelegraph.com/news/fake-spacex-youtube-channels-...

The first one I saw was a Rip of a SpaceX livestream. I watched it for a bit then noticed all the BitCoin references and got confused then realized it was a scam account. How they get promoted basically to the front page is the issue.

A solution as simple as just hiring someone for $10/hour to periodically scan youtube for crypto keywords and disable the scam videos and accounts would have prevented millions of dollars of theft.

I think this is a good solution, but maybe not viable. Going past Crypto this could apply to so many issues needed to be reviewed on YouTube. But at that scale, the human cost is something YouTube refuses to pay.

it would be a stopgap measure until the algos are refined to filter out the videos without the need for human intervention. YouTube is being sued by Brad Garlinghouse for not taking action fast enough. My guess is my $10/hour solution is cheaper than $1000/hour lawyers.

Those lawyers already spent the money writing a ToS that says they're not liable for shit.

What are examples of keywords that would have a low false positive (and ideally, low false negative) rate, such that a minimum wage worker could efficiently disable videos with little context or time?

it is not the keywords but the content of the videos. anything that involves a livestream and a pitch to send X to get 2x-10x back. very easy to train someone to identify what these scam look like with zero low false positive rate

Why would you need a human to do a job that a computer can do 100x faster?

> "What are examples of keywords that would have a low false positive"

"Bitcoin"

Reply on news.ycombinator.com