Agreed. We're becoming more and more a certificate-society than one based on true understanding. There are few mechanisms used to evaluate people's abilities that aren't based on whether or not you have a piece of paper.
“When you bet on a winner, diversifying is likely to be a case of you selling the winner to buy the loser.”
Well yeah, that assumes you know the winner. The point of diversifying is that you don’t know the winner. So, it’s better to make a few guess with a chance of making up losses than likely backing a total loser.
Apple began scanning for CSAM with Neuralhash. This allows you to turn an image into a specific neuralhash thus possibly triggering its (automatic) CSAM detection. Imagine if a picture of a cat could cause Apple to think you have CP on your device.
Well, you'd have to do it 30 times to trigger the system, and then someone at apple moderation would look at those 30 pictures of cats and hit "next" vs "supervisor"
Good that there’s some human supervision. But, I know I have more than 30 photos of my dog. Also don’t like the idea of false positives auto-sharing some of my camera roll.
It's only if you back it up to iCloud, the signatures of the CP used as references are rotated, and they're also not public. The chances of you randomly triggering the system is effectively 0 unless you're uploading CP to your iCloud.
No. They calculate a hash on the device, but they only do it as part of the iCloud upload. So whether the hashing happens on the device or on the server, the same images get hashed either way.
Photos of your dog are not going to trigger it. Someone would need to engineer the 30 photos of your dog tweaked to hash to a particular value, and then convince you to save them to your device and then upload to iCloud. And then some portion/abstraction of the dog photo would need to convince a reviewer they were looking at CSAM.
The more likely path to trouble is legal NSFW material that's been engineered.
You can be pretty sure they'll report your account if at least 1 low-res thumbnail ("visual derivative") looks like an image of naked people/a sexual act.
You're not trying hard enough on how to bypass the human component methinks.
Use porn as the base images. The more petite, flat and young looking, the better. The moderators are already going to be tuned in to csam, so all you need to do is to give them a slight push.
Just pushed my first public GitHub repo! It's a simple password manager in Python for the terminal. I thought it was about time I started “building in public” and learning more about encryption.
This process gave me a serious appreciation for what has to be done to protect your data.
Would like to emphasize this. They’re some great tools that manage knowledge extremely well (including Innos). But now I’d like to see more work on the knowledge’s deployment into the user’s life — a tool to help make it actionable (in conversation for example) without having to boot up the library every time.
I see that Innos has a programming feature which I think is a good step. But more features that specifically help the recall of knowledge would be awesome.
