[ExI] 23andme again

Mike Dougherty msd001 at gmail.com
Thu Jul 4 01:12:00 UTC 2013


On Wed, Jul 3, 2013 at 6:40 PM, spike <spike at rainier66.com> wrote:
>>...Suppose I swipe a copy of spike's gene sequence directly from 23andMe
> computers.  Then I get a 3D printer, DIY bio hacking genetic materials
> cooker, and now I have a patsy to take the blame for whatever crime I want
> to place this material.  Ok sure, I don't have an complete match for spike -
> but i'm not trying to clone anyone, just provide a conclusive enough match
> against the database to create a fraudulent positive ID...
>
> Good that you mention that Michael.  We need some kind of system to prevent
> this kind of thing.  If my anticipated development actually occurs, that
> medical insurance groups will begin cherry-picking based on DNA profiles,
> then we need some means of preventing the insurance companies from inserting
> signals into the genome file to convince proles they are at risk, so they
> should pay more.  The insurance companies will take care of themselves to
> prevent risk-markery people from borrowing the genome files of the
> non-risk-markery people to get their insurance cheaper.

As I continued to think about my first proposal for incriminating the
unsuspecting 23andSpike I realized that if I could hack the computer,
it would be far easier to simply swap a few records in the database.
Now my genetics point to your information and I've succeeded in a kind
of identity theft.  In this case, there's probably not as much
monitoring for integrity on this data so I'm probably going to get
away with it for longer than stealing your credit card(s).  I wonder
if it would be sufficiently confusing to swap my profile in several
databases so the cross-match from 23andMe to NSA's DB to Google/Apple
DB, etc. would be strangely inaccurate.  Of course if only my own was
incorrect it could be discerned and fixed.  I'll be sure to swap a few
thousand records in several disconnected sets/graphs.  At least I
imagine this is easier than the actual biological parts of creating
DNA from computer files... but I'm no DIY geneticist.  There may be a
point where hobby genome creation is no more expensive than 3D
printing.  How this will impact laws should be interesting.

>>...As clever as you think you will be catching the crooks, plan on the
> crooks getting more clever too...
>
> Ja.  The term crooks vaguely implies those who wish to steal.  I don't care
> that much about catching them.  The ones I want is to catch are the violent
> criminals and rapists.  I don't really perceive that set as getting any
> smarter over time; if anything perhaps dumber.  Those who would steal by
> fraud, ja, definitely getting smarter.

ok, crooks may not have implied the full scope of my intent.  If I
wanted to destroy your public image, personal brand, career
opportunity, or political aspirations I need only subject your DNA to
a host of paternity claims (you untrustworthy philanderer) or place
your DNA at the scene of a horrendous crime (which I would have to
outsource to some thugs)

I appreciate that DNA is currently a highly accurate form of biometric
identification.  The systems we use to track and match these
biological signatures are hardly foolproof.  Whether this will ever be
explained to the proles remains to be discovered.



More information about the extropy-chat mailing list