Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ImageNet contains naturally occurring Apple NeuralHash collisions

blog.roboflow.com

Post reply on HN
Print view View original

91–100 of 530 posts

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#91

Post by smithza » 

     This is a false-positive rate of 2 in 2 trillion image pairs (1,431,168^2). Assuming the NCMEC database has more than 20,000 images, this represents a slightly higher rate than Apple had previously reported. But, assuming there are less than a million images in the dataset, it's probably in the right ballpark.
If the author was comparing 2 trillion pictures of people, or children specifically, I think this false-positive rate would be different and arguably much higher. The reasons are obvious: humans are similar in dimensions to eachother and are much more likely to match in the same way the hatchet and nematode matched.

I do not presume such a finding of photos is easy to come by but I wish the author put details on the sample set.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#92

Post by yeldarb »

slg wrote: post #48
bawolff wrote: post #9

Earlier quoted context omitted.

ImageNet is a very well-known data set. Are we sure apple didn't test on it when designing this algorithm?

>This is a false-positive rate of 2 in 2 trillion image pairs (1,431,168^2). Assuming the NCMEC database has more than 20,000 images, this represents a slightly higher rate than Apple had previously reported. But, assuming there are less than a million images in the dataset, it's probably in the right ballpark. Apple reported a pretty similar collision rate so maybe they did.

Their reported collision rate was against their CSAM hash database[1].

> In Apple’s tests against 100 million non-CSAM images, it encountered 3 false positives when compared against NCMEC’s database. In a separate test of 500,000 adult pornography images matched against NCMEC’s database, it found no false positives.

[1] https://tidbits.com/2021/08/13/new-csam-detection-details-em...

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#93

Post by nonbirithm »

dannyobrien wrote: post #64
endisneigh wrote: post #29

Earlier quoted context omitted.

Ok sure, PhotoDNA has existed since 2008. Where are the instances of people being sent to prison?

One of the things that is happening now is that the entire PhotoDNA system is finally coming under the level of oversight that it should have had right from the start. I can tell you from working in this area that it's possible for someone to have their lives ruined by a misplaced investigation, have that investigation abandoned because they turn out to be obviously innocent, and for that to not be well-known, becaus…

We certainly have Apple's PR blunder to thank for that, but if PhotoDNA always held that potential for abuse due to its very nature, why did we remain silent for 13 years?

Maybe it's because Google and Microsoft and others' policy of security through obscurity actually succeeded in preventing the details of PhotoDNA from coming to light, and it took Apple exposing their hashing model to reverse engineering by including it on the device for people to finally wake up.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#94

Post by beepbooptheory »

stickfigure wrote: post #19

> it's not obvious how we can trust that a rogue actor (like a foreign government) couldn't add non-CSAM hashes to the list to root out human rights advocates or political rivals. Apple has tried to mitigate this by requiring two countries to agree to add a file to the list, but the process for this seems opaque and ripe for abuse. If the CCP says "put these hashes in your database or we will halt all iPhone sales in…

The sinophobia of this website is getting to be more than tiring and really just becoming worrisome.

Would never of bet that such smart and thoughtful people in general could be so so stuck within certain political talking points, to an almost fanatical degree. If people on HN are generally like this, I have little hope for the rest of our world and its future.

At the end of the day all the bad guys will still win, even if we did give a lot of attention to a subset of them for a while I guess.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#95

Post by rootusrootus »

Dah00n wrote: post #50
endisneigh wrote: post #29

Earlier quoted context omitted.

Ok sure, PhotoDNA has existed since 2008. Where are the instances of people being sent to prison?

You are commenting a lot for this many places in the thread. Are you arguing for this system or for Apple? It reads like pro-Apple and doesn't add anything except "I think it is good, therefore it is good".

It can get a little frustrating to hear so much inaccurate FUD being spread around which detracts from a reasoned discussion on the merits.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#96

Post by nybble41 »

tehnub wrote: post #51
ulzeraj wrote: post #25

Earlier quoted context omitted.

What if those two countries can be Poland and Hungary? These two countries have been passing lots of laws to ostracize and criminalize pro-LGBT content and are friendly to each other.

Apple did mention in their security thread model document [0] this: Apple will also refuse all requests to instruct human reviewers to file reports for anything other than CSAM materials for accounts that exceed the match threshold. [0]: https://www.apple.com/child-safety/pdf/Security_Threat_Model...

That's too little, too late. By the time any human reviewer can evaluate whether the images should have been in the database the encryption on the backups has already been circumvented and other parties have already been given access to your private files.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#97

Post by qweqwweqwe-90i »

Dah00n wrote: post #50
endisneigh wrote: post #29

Earlier quoted context omitted.

Ok sure, PhotoDNA has existed since 2008. Where are the instances of people being sent to prison?

You are commenting a lot for this many places in the thread. Are you arguing for this system or for Apple? It reads like pro-Apple and doesn't add anything except "I think it is good, therefore it is good".

FYI: when you wrote this comment, you had posted 5% of all the comments on this thread.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#98

Post by yeldarb »

smithza wrote: post #91

This is a false-positive rate of 2 in 2 trillion image pairs (1,431,168^2). Assuming the NCMEC database has more than 20,000 images, this represents a slightly higher rate than Apple had previously reported. But, assuming there are less than a million images in the dataset, it's probably in the right ballpark. If the author was comparing 2 trillion pictures of people, or children specifically, I think this false-posi…

The sample set is ImageNet, which is a well-known dataset in Computer Vision and is available for download here: https://www.kaggle.com/c/imagenet-object-localization-challe...

I'd love to see this work extended; if you find additional collisions in the wild please submit a PR to the repo (please do not submit artificially generated adversarial images): https://github.com/roboflow-ai/neuralhash-collisions

For what it's worth, Apple claimed to find a _lower_ incidence of false-positives when it used pornographic images in its test[1] (which makes sense; images containing humans is probably more aligned with what the model was trained on than nematodes)

[1] https://tidbits.com/2021/08/13/new-csam-detection-details-em...

> In Apple’s tests against 100 million non-CSAM images, it encountered 3 false positives when compared against NCMEC’s database. In a separate test of 500,000 adult pornography images matched against NCMEC’s database, it found no false positives.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#99

Post by tehnub »

sweezyjeezy wrote: post #72
tehnub wrote: post #47

> In order to test things, I decided to search the publicly available ImageNet dataset for collisions between semantically different images. I generated NeuralHashes for all 1.43 million images and searched for organic collisions. By taking advantage of the birthday paradox, and a collision search algorithm that let me search in n(log n) time instead of the naive n^2, I was able to compare the NeuralHashes of over 2…

By "taking advantage of the Birthday Paradox" - he means that even though the chance of two random images colliding is ~ 1/1 trillion, if you have a set of size ~ sqrt(1 trillion) you have a good chance of having a collision amongst all pairs.

But did the birthday paradox inform his procedure in any way? I guess he wouldn't have even attempted to do this at all if he didn't think it was likely to find collisions.

Re: ImageNet contains naturally occurring Apple NeuralHash collisions

#100

Post by kemayo »

Dah00n wrote: post #50
endisneigh wrote: post #29

Earlier quoted context omitted.

Ok sure, PhotoDNA has existed since 2008. Where are the instances of people being sent to prison?

You are commenting a lot for this many places in the thread. Are you arguing for this system or for Apple? It reads like pro-Apple and doesn't add anything except "I think it is good, therefore it is good".

If you have a point which you feel rebuts a common argument, it seems reasonable to leave that comment in places you see that argument. The alternative is "minority positions should be drowned out", no?
Post reply on HN