Houston man charged with child porn possession after Google cyber-tip

Discussion in 'Politics, Religion, Social Issues' started by Dolorian, Aug 4, 2014.

  1. Dolorian, Aug 4, 2014
    Last edited: Aug 4, 2014

    Dolorian macrumors 65816

    Dolorian

    Joined:
    Apr 25, 2007
    #1
    This news item has been making the rounds lately...

    Source

    ...

    So, how did Google know it was a CP image?. Taking this from another comment in this link:

    Google were sent the 'fingerprint' identifiers of known child sex images by the Child Victim Identification Program. They then checked to see if any of these matched images being sent by Gmail users. Once they had a potential match, the advised the police who then were able to obtain a warrant to search the account. The search proved positive and the paedophile was caught.

    This isn't Google checking to see what all of our photos are to make sure they're 'acceptable', it's Google asking their database if any images match known illegal image fingerprints, and then alerting the police if there's a match.

    It's like the post service being told by the CIA 'We know letter-bombs are being sent in packages that look like these:' Should the post service flag any packages to the police they notice that match the description of the letter bombs? Or should they deliver them because they should mind their own business, despite the fact they're leaving the opening of the packages to the police?

    ...

    Sounds reasonable to me.
     
  2. Tomorrow macrumors 604

    Tomorrow

    Joined:
    Mar 2, 2008
    Location:
    Always a day away
    #2
    No, it's more like the post service looking inside your packages to see if there's anything in there the CIA is interested in.

    I understand what you're getting at, but there's a difference between looking at the outside of a package and looking inside of it; same as there's a difference between looking at an email header and its content.
     
  3. aerok macrumors 65816

    aerok

    Joined:
    Oct 29, 2011
    #3
    Very reasonable in my opinion too. And I'm sure for those that receive emails with illegal pictures by mistake or SPAM are safe unless they share the pictures.
     
  4. lannister80 macrumors 6502

    lannister80

    Joined:
    Apr 7, 2009
    Location:
    Chicagoland
    #4
    BUT, if it's not something the CIA is interested in, they don't know what it is.

    Image hashes are basically a magic wand that is always right and never wrong when detecting image "contraband". And if it doesn't match the hash, the algorithm moves on, having revealed nothing about the non-matching image (except the hash).

    Of course, that assumes the list of hashes they're checking for is a legit list in the first place. AND that the person whose account got tagged isn't being set up by someone else.
     
  5. Sydde macrumors 68020

    Sydde

    Joined:
    Aug 17, 2009
    #5
    From what I can tell, they did not examine the image itself but looked at its meta-data (exif-type stuff) and saw something questionable. It is more akin to a bomb-sniffing dog that a look-see or X-ray. And the cops got a warrant.

    Now, obviously, just a slight alteration with GC or PS might be enough to scrub the image of the metadata. Fortunately, most of these kind of people are not that clever.
     

Share This Page