[Full-disclosure] OMIGOD CIQ HACKING THE WORLD.
dan.j.rosenberg at gmail.com
Wed Dec 7 13:43:40 GMT 2011
And I was really hoping I wouldn't get dragged into another discussion
On Wed, Dec 7, 2011 at 7:55 AM, Pablo Ximenes <pablo at ximen.es> wrote:
> Hi All,
> Based on what I read from the post, basically Rosenberg recognises he has no
> clue about what happens with the rest of affected phone models:
> "One important thing to note is that this represents the metrics that are
> submitted to the CarrierIQ application by the code written by Samsung. The
> list of available metrics are carrier specific, but will remain constant on
> a given handset model. The subset of this data that is actually recorded and
> collected is at the discretion of the carrier, and is based on the profile
> installed on the device." (Dan Rosenberg)
> So the eavesdropped data with respect to the rest of affected phones could
> be anything for all he knows, including contents of SMS's and visited pages.
This is not accurate. I have not enumerated the locations on every
phone where OEMs have modified the Android framework to submit metrics
to the CarrierIQ agent running on the phone. This means I don't know
what subset of metrics that CIQ supports is actually being submitted
to the agent on every phone. However, I have reverse engineered the
actual CarrierIQ binaries on a wide variety of phones (the agent code
is actually fairly similar across the board), so I have a very good
idea of the total set of supported metrics (regardless of what's
actually being submitted) looks like. And surprise, there is no
metric that contains fields for SMS bodies or web page contents.
Food for thought: why would a carrier double their bandwidth
utilization for SMS in order to violate federal wiretapping law and
get a second copy of the text message *that they already have*?
> And about collecting every URL (even https ones) that is visited. Forget
> about the legality, let's go directly to the privacy implications.
> For instance, if you do that for a simple Facebook session, there's a huge
> amount of very private information being collected (fixed URLS that reveal
> photos, etc; ajax URLs that reveal juicy IDs, among other things). Also, I
> don't think anybody would want to have their complete web history in the
> hands of anyone without their express consent.
Agreed. It will be interesting to see whether or not this information
is actually being collected, since I've only shown that it's
*possible* for it to be collected, not that it actually is.
> Going back to the legality, even if the URL is just the begining of an HTTP
> negotiation process, it doesn't mean that URLs are not payloads legally. In
> many countries only layers 4 (transport) and bellow (TCP info, IP data, etc)
> would be considered header information and all the rest would be considered
> payload, incluing the URL. If what Rosenberg claims is that a URL is not
> considered payload to the law, I thing he might have to review his concepts.
> In Brazil, for instance, capturing the URL alone in this scenario would
> constitute a crime of illegal wiretapping.
I never made this claim. I explicitly state that the legality of this
needs to be investigated, since to my knowledge, it's an open legal
question in the United States.
> Pablo Ximenes
> 2011/12/6 Christian Sciberras <uuf6429 at gmail.com>
>> Or not...
>> On the other hand, where that l33t hacker Drew (aka xD 0x41)?
>> Thought he'd enlighten us with more of his awesome hacking powers on this
>> Full-Disclosure - We believe in it.
>> Charter: http://lists.grok.org.uk/full-disclosure-charter.html
>> Hosted and sponsored by Secunia - http://secunia.com/
> Full-Disclosure - We believe in it.
> Charter: http://lists.grok.org.uk/full-disclosure-charter.html
> Hosted and sponsored by Secunia - http://secunia.com/
Full-Disclosure is hosted and sponsored by Secunia.