This is some "don't-worry-ism" if I ever saw some. Hopemongering, arguably. The stats are pretty clear: mental illness is going up. [1] [2]

[1] https://www.insuranceproviders.com/best-and-worst-cities-for...

[2] https://www.thecut.com/2016/03/for-80-years-young-americans-...

I actually think modern society is horrible in all sorts of ways leading to, among other things, rising rates of mental health issues. But I fail to see the connection between rising rates of depression and mental health issues and what's being discussed here.

Diagnosis of depression going up (as shown in the first link) don't necessarily mean incidents of depression are going up. It could be that more people are trying to treat it than before.

The stats are clear. Causality... not so much.

Aye, we're diagnosing things at a much higher rate than before.

I am not sure how this relates to OPs comment about "f*cked up" videos being posted. There is no connection between depression and becoming interested in posting gruesome videos.

It wouldn't be 10 million videos/month, it would be 10 million active users. This is a notable difference because each of these users can be uploading numerous content and tiktok users who have posted content are much more likely to continue posting content far more frequently than just one video per month.

Yes, it was just a simplistic analysis to demonstrate a point about large numbers of people.

Theoretically it now does not seem too difficult that data science could actually identify who these people are based on the amount of metadata we all generate. Are there attempts to be preemptive in this regard?

The same kind of "data science" that identified many people and businesses named after the Egyptian goddess with an Islamic terror group and took automatic action against them?

I think we need to move to some way to identify actual people on the internet. I don't mean that in a "use your real name and upload your passport"-kind of way, but in a "we can 100% reliably tell you've been a twat in the past, whomever you are, so no new signups for you"-kind of way that also accounts for all the various interests such as right to privacy and anonymity.

Many platforms use phone numbers for that now, which is obviously far from perfect as I can easily get 100 of those today if I wanted to.

That's not a ludicrous idea. India requires activity be logged to a specific individual, for instance. Seems a bit dystopian for my sensibilities though.

I'm sure there are, but do we want there to be?

https://en.wikipedia.org/wiki/Minority_Report_(film)

It doesn’t have to be that apocalyptic. Take the literal traffic enforcement for example. Driving faster than the speed limit gets traffic fines. Cities have cameras for red light or toll booth violations and have databases of offenders.

"Databases of offenders" is different from "database of people who haven't yet done anything but statistically we think might". I could be reading it wrong, but it sounded to me like the parent comment was talking about predicting people who might post such videos _before_ they did. I don't think you'd need to theorize about using data science to ban people after they post some number of violating videos.

Social media companies can be passive, reactive, proactive, or predictive in their efforts to ban bad actors. How much effort are they putting in at each level? It seems more towards the least effort.

Just because we can imagine it going wrong doesn't mean it will go wrong or it must go wrong. The role of science fiction is to explore possible futures, not to make the future taboo.

Do we need to be way more careful with how we design our society? Yes. Does it make sense to abandon possible solutions because of a movie? No.

>I don't know how accurate that 1 billion number of users is exactly

Ask Elon Musk to offer to buy it, then we'll see how accurate those numbers are.

In reality, I never believe people's published numbers. OF COURSE they are inflated. Their entire (maybe majority?) valuations are dependent on those numbers being as high as possible.

I didn't even check the source of those numbers; it was just the first thing that came up in a very quick search. It doesn't really matter: even with 500 million or 100 million you've got the same "a small percentage of a very large group can cause a lot of havoc"-effect. You see the same if you drive to work: you can easily encounter over a 100 of other drivers, and only a very small percentage of them can give the perception that "people are such bad drivers" when in fact it's just the 1% of the 100 people you encountered on that trip.

I find it both depressing and comforting; it's comforting to know most people aren't that bad, and it's depressing that such a small group of assholes can fuck it all up.

You might be off by another factor of 100 because most of those "active users" just consume stuff and maybe add a comment.

https://en.m.wikipedia.org/wiki/1%25_rule

It doesn’t even matter — going back to the car example, say you drive near 1-2 thousand cars on an hour drive, if you have 2-3 scary experiences, that makes you think “drivers here suck!” So a sample size of like .1% makes you think a problem applies to 50% of drivers.

If I watch TikTok for an hour a day for a week, and I watch 4 videos/minute on average, I’ll watch over 1500 videos a week. If only one or two of those are VERY traumatizing (e.g. showing death), I’d think TikTok has a big moderation problem. That’s an ultra-low number of users making it seem like there is a huge issue

Disclaimer: I don’t have TikTok

TikTok having an estimated 1bn users is quite low when you compare it to the likes of Google's YouTube at twice that (2.5bn). If anything I expected TikTok to have passed the 2bn mark already considering its reach, most notably surpassed by Meta's subsidiaries and YouTube.

Of course now we have to have the conversation of "what is a user". ADU, AMU, people with app installed, etc.

If is banned in India and not used in China, so that limits reach.

> OF COURSE they are inflated

Bytedance is pre-IPO. However, you are accusing Google, Facebook, snapchat, Twitter of securities fraud. What is the latest from Musk, on Twitter's alleged securities fraud?

I'm not accusing anyone of anything other than padding their numbers. How you interpret that is up to you.

Do you really believe TikTok as exactly 1 Billion users, or 999,990,000+ users? Anyone that says that is 1 Billion is padding the numbers. Nobody is going to call that 999 million.

If the number is anything close to being rounded, the number is padded. Unless they are showing you an exact number from the time the count was read, it is being padded.

I think (I'm hoping?) 1% figure is vastly overestimated.