Alicorn468 wrote:

Scratch could use an AI that could get smarter by each pfp that gets into scratch but idk

Machine learning is way too complex.

Alicorn468 wrote:

Scratch could use an AI that could get smarter by each pfp that gets into scratch but idk

Machine learning is way too complex.

Why did it post twice?

NecroAnvil wrote:

If this were to be implemented, people would try to find out how it works and when they do, try to workaround it.

Why are you so against it?
Minimizing it is better then nothing and younger scratchers shouldn’t have to see it when there could be a pfp filter.reporting can take hours.and in the hours more people have to see it.not everyone would be able to workaround it and they’d have to hack into scratch.if there’s a filter they get banned sooner or at least suspended.it might not happen as much even if some gets threw.

Alicorn468 wrote:

Miiraaya wrote:

I've seen people trying to break the community guidelines by using inappropriate icons, however I think there should be a system for blocking out the aforementioned (i.e. pictures of faces, 18+ content, e.t.c.) which would make Scratch a more kid-friendly site

Scratch could use an AI that could get smarter by each pfp that gets into scratch but idk

It's made by pixel scanning.

By the way, I BARELY support. I have seen many bad-for-scratch PFPs, and the Scratch Team really needs to do something about that.

On the other side, the filter can be inaccurate a lot of the times. So… I'm neutral.

musicROCKS013 wrote:

Why did it post twice?

It's a common glitch in Scratch nowadays.

Alicorn468 wrote:

Scratch could use an AI that could get smarter by each pfp that gets into scratch but idk

Machine learning is way too complex.

It would be really difficult to create an AI that can filter out bad PFPs. The Scratch Team devs haven't even had the chance to finish the rest of the Scratch 3 styled site, so obviously wouldn't have enought time and resources to create an AI to filter PFPs and train it too.

kccuber wrote:

It would be really difficult to create an AI that can filter out bad PFPs. The Scratch Team devs haven't even had the chance to finish the rest of the Scratch 3 styled site, so obviously wouldn't have enought time and resources to create an AI to filter PFPs and train it too.

There are already programs that filter nsfw content.

CaIIMeHoid wrote:

TheAnimalKingdom42 wrote:

How would the filter know when a profile picture was 18+?

I'd imagine something like this. Your text to link here…

Exactly what I thought

Then don’t have face detection.have google images of 18+ stuff implemented without them having too look threw them all.

Prince_Wolf1 wrote:

NecroAnvil wrote:

If this were to be implemented, people would try to find out how it works and when they do, try to workaround it.

Why are you so against it?

They are not against it, they are just sharing their opinions.

starhero5697 wrote:

Prince_Wolf1 wrote:

NecroAnvil wrote:

If this were to be implemented, people would try to find out how it works and when they do, try to workaround it.

Why are you so against it?

They are not against it, they are just sharing their opinions.

Hmm… That's a bit sus…

With that out of the way, let me bring up a post I made a page ago.

k0d3rrr wrote:

TheAnimalKingdom42 wrote:

How would the filter know when a profile picture was 18+?

Yeah. And even then, how accurate would the filter be?
It might censor a completely innocent PFP, while it might allow a completely inappropriate PFP.

starhero5697 wrote:

Prince_Wolf1 wrote:

NecroAnvil wrote:

If this were to be implemented, people would try to find out how it works and when they do, try to workaround it.

Why are you so against it?

They are not against it, they are just sharing their opinions.

They’ve been against it.
We accidentally made a duplicate and they were there and were against it.

k0d3rrr wrote:

TheAnimalKingdom42 wrote:

How would the filter know when a profile picture was 18+?

Yeah. And even then, how accurate would the filter be?
It might censor a completely innocent PFP, while it might allow a completely inappropriate PFP.

Well it doesn’t need to sensor faces, actual pictures of peoples faces doesn’t seem very common, unless it faces of famous people

Couldn’t you just report the photo?

CherryBlossoms-- wrote:

Couldn’t you just report the photo?

Yes we can and do but it takes six or more hours before they get banned and they just make another account. If we can prevent we should prevention is better then just reporting the..

Prince_Wolf1 wrote:

(#32)

starhero5697 wrote:

Prince_Wolf1 wrote:

NecroAnvil wrote:

If this were to be implemented, people would try to find out how it works and when they do, try to workaround it.

Why are you so against it?

They are not against it, they are just sharing their opinions.

They’ve been against it.
We accidentally made a duplicate and they were there and were against it.

Still, they were just sharing their opinions. Don't jump to conclusions, Besides, what's wrong with them being against the suggestion? Are you trying to make them support? Wait, do you even know how the support/no support convention works?

Guys, guys! We're getting off-topic here! Quit bothering ‘Anvil and ’Star, okay?

Anyways, I don't know how to feel about this suggestion. While I think it could be great having a moderation system for PFPs, I truly think that it wouldn't be very accurate. It could block PFPs that have nothing wrong with them.

Another problem I can see with a filter system is that people could still share their phone number on their PFP without it getting blocked (don't) because the filter isn't smart enough to think that that's a phone number, and that all it sees is a string of numbers in a row.

k0d3rrr wrote:

(#37)
Another problem I can see with a filter system is that people could still share their phone number on their PFP without it getting blocked (don't) because the filter isn't smart enough to think that that's a phone number, and that all it sees is a string of numbers in a row.

In comments, there is already a filter for phone numbers, maybe if the PFP filter sees a string of numbers, they could censor it?

starhero5697 wrote:

In comments, there is already a filter for phone numbers, maybe if the PFP filter sees a string of numbers, they could censor it?

This shouldn’t be a thing. It would result in many false positives. For example, k0d3rrr’s pfp has his username in it, and it would detect those words and block him from using that pfp.

I would suggest that it could identify the letters, and if it suspects that it is a phone number it will be reviewed by a moderator.

Nvm

starhero5697 wrote:

k0d3rrr wrote:

(#37)
Another problem I can see with a filter system is that people could still share their phone number on their PFP without it getting blocked (don't) because the filter isn't smart enough to think that that's a phone number, and that all it sees is a string of numbers in a row.

In comments, there is already a filter for phone numbers, maybe if the PFP filter sees a string of numbers, they could censor it?

Yeah, that would make sense.
GOT NINJA'D BY FOUR SECONDS LOL.