[Paleopsych] WP: A Trail of DNA and Data

Premise Checker checker at panix.com
Mon Apr 4 17:17:59 UTC 2005


A Trail of DNA and Data
http://www.washingtonpost.com/ac2/wp-dyn/A20454-2005Apr2?
5.4.3
    By Paul Saffo

    If you're worried about privacy and identity theft, imagine this:

    The scene: Somewhere in Washington. The date: April 3, 2020.

    You sit steaming while the officer hops off his electric cycle and
    walks up to the car window. "You realize that you ran that red light
    again, don't you, Mr. Witherspoon?" It's no surprise that he knows
    your name; the intersection camera scanned your license plate and your
    guilty face, and matched both in the DMV database. The cop had the
    full scoop before you rolled to a stop.

    "I know, I know, but the sun was in my eyes," you plead as you fumble
    for your driver's license.

    "Oh, don't bother with that," the officer replies, waving off the
    license while squinting at his hand-held scanner. Of course. Even
    though the old state licensing system had been revamped back in 2014
    into a "secure" national program, the new licenses had been so
    compromised that the street price of a phony card in Tijuana had
    plummeted to five euros. In frustration, law enforcement was turning
    to pure biometrics.

    "Could you lick this please?" the officer asks, passing you a
    nanofiber blotter. You comply and then slide the blotter into the
    palm-sized gizmo he is holding, which reads your DNA and runs a match
    against a national genomic database maintained by a consortium of drug
    companies and credit agencies. It also checks half a dozen metabolic
    fractions looking for everything from drugs and alcohol to lack of
    sleep.

    The officer looks at the screen, and frowns, "Okay, I'll let you off
    with a warning, but you really need more sleep. I also see that your
    retinal implants are past warranty, and your car tells me that you are
    six months overdue on its navigation firmware upgrade. You really need
    to take care of both or next time it's a ticket."

    This creepy scenario is all too plausible. The technologies described
    are already being developed for industrial and medical applications,
    and the steadily dropping cost and size of such systems will make them
    affordable and practical police tools well before 2020. The resulting
    intrusiveness would make today's system of search warrants and
    wiretaps quaint anachronisms.

    Some people find this future alluring and believe that it holds out
    the promise of using sophisticated ID techniques to catch everyone
    from careless drivers to bomb-toting terrorists in a biometric
    dragnet. We have already seen places such as Truro, Mass., Baton
    Rouge, La. and Miami ask hundreds or thousands of citizens to submit
    to DNA mass-testing to catch killers. Biometric devices sensing for
    SARS symptoms are omnipresent in Asian airports. And the first
    prototypes of systems that test in real time for SARS, HIV and bird
    flu have been deployed abroad.

    The ubiquitous collection and use of biometric information may be
    inevitable, but the notion that it can deliver reliable, theft-proof
    evidence of identity is pure science fiction. Consider that oldest of
    biometric identifiers -- fingerprints. Long the exclusive domain of
    government databases and FBI agents who dust for prints at crime
    scenes, fingerprints are now being used by electronic print readers on
    everything from ATMs to laptops. Sticking your finger on a sensor
    beats having to remember a password or toting an easily lost smart
    card.

    But be careful what you touch, because you are leaving your identity
    behind every time you take a drink. A Japanese cryptographer has
    demonstrated how, with a bit of gummi bear gelatin, some cyanoacrylic
    glue, a digital camera and a bit of digital fiddling, he can easily
    capture a print off a glass and confect an artificial finger that
    foils fingerprint readers with an 80 percent success rate. Frightening
    as this is, at least the stunt is far less grisly than the tale,
    perhaps aprocryphal, of some South African crooks who snipped the
    finger off an elderly retiree, rushed her still-warm digit down to a
    government ATM, stuck it on the print reader and collected the
    victim's pension payment. (Scanners there now gauge a finger's
    temperature, too.)

    Today's biometric advances are the stuff of tomorrow's hackers and
    clever crooks, and anything that can be detected eventually will be
    counterfeited. Iris scanners are gaining in popularity in the
    corporate world, exploiting the fact that human iris patterns are
    apparently as unique as fingerprints. And unlike prints, iris images
    aren't left behind every time someone gets a latte at Starbucks. But
    hide something valuable enough behind a door protected by an iris
    scanner, and I guarantee that someone will figure out how to capture
    an iris image and transfer it to a contact lens good enough to fool
    the readers. And capturing your iris may not even require sticking a
    digital camera in your face -- after all, verification requires that
    the representation of your iris exist as a cloud of binary bits of
    data somewhere in cyberspace, open to being hacked, copied, stolen and
    downloaded. The more complex the system, the greater the likelihood
    that there are flaws that crooks can exploit.

    DNA is the gold standard of biometrics, but even DNA starts to look
    like fool's gold under close inspection. With a bit of discipline, one
    can keep a card safe or a PIN secret, but if your DNA becomes your
    identity, you are sharing your secret with the world every time you
    sneeze or touch something. The novelist Scott Turow has already
    written about a hapless sap framed for a murder by an angry spouse who
    spreads his DNA at the scene of a killing.

    The potential for DNA identity theft is enough to make us all wear a
    gauze mask and keep our hands in our pockets. DNA can of course be
    easily copied -- after all, its architecture is designed for
    duplication -- but that is the least of its problems. Unlike a credit
    card number, DNA can't be retired and swapped for a new sequence if it
    falls into the hands of crooks or snoops. Once your DNA identity is
    stolen, you live with the consequences forever.

    This hasn't stopped innovators from using DNA as an indicator of
    authenticity. The artist Thomas Kinkade signs his most valuable
    paintings with an ink containing a bit of his DNA. (He calls it a
    "forgery-proof DNA Matrix signature.") We don't know how much of Tom
    is really in his paintings, but perhaps it's enough for forgers to
    duplicate the ink, as well as the distinctive brush strokes.

    The biggest problem with DNA is that it says so much more about us
    than an arbitrary serial number does. Give up your Social Security
    number and a stranger can inspect your credit rating. But surrender
    your DNA and a snoop can discover your innermost genetic secrets --
    your ancestry, genetic defects and predispositions to certain
    diseases. Of course we will have strong genetic privacy laws, but
    those laws will allow consumers to "voluntarily" surrender their
    information in the course of applying for work or pleading for health
    care. A genetic marketplace not unlike today's consumer information
    business will emerge, swarming with health insurers attempting to
    prune out risky individuals, drug companies seeking customers and
    employers managing potential worker injury liability.

    Faced with this prospect, any sensible privacy maven would conclude
    that DNA is too dangerous to collect, much less use for a task as
    unimportant as turning on a laptop or working a cash machine. But
    society will not be able to resist its use. The pharmaceutical
    industry will need our DNA to concoct customized wonder drugs that
    will fix everything from high cholesterol to halitosis. And crime
    fighters will make giving DNA information part of our civic duty and
    national security. Once they start collecting, the temptation to use
    it for other purposes will be too great.

    Moreover, snoops won't even need a bit of actual DNA to invade our
    privacy because it will be so much easier to access its digital
    representation on any number of databanks off in cyberspace. Our Mr.
    Witherspoon will get junk mail about obscure medical conditions that
    he's never heard of because some direct marketing firm "bot" will
    inspect his digital DNA and discover that he has a latent disease or
    condition that his doctor didn't notice at his annual checkup.

    It is tempting to conclude that Americans will rise up in revolt, but
    experience suggests otherwise. Americans profess a concern for
    privacy, but they happily reveal their deepest financial and personal
    secrets for a free magazine subscription or cheesy electronic trinket.
    So they probably will eagerly surrender their biometric identities as
    well, trading fingerprint IDs for frequent shopper privileges at the
    local supermarket and genetic data to find out how to have the
    cholesterol count of a teenager.

    Biometric identity systems are inevitable, but they are no silver
    bullet when it comes to identity protection. The solution to identity
    protection lies in the hard work of implementing system-wide and
    nationwide technical and policy changes. Without those changes, the
    deployment of biometric sensors will merely increase the opportunities
    for snoops and thieves -- and escalate the cost to ordinary citizens.

    It's time to fix the problems in our current systems and try to
    anticipate the unique challenges that will accompany the expanded use
    of biometrics. It's the only way to keep tomorrow's crooks from
    stealing your fingers and face and, with them, your entire identity.

    Paul Saffo is a director of the Institute for the Future, a
    forecasting organization based in Silicon Valley.



More information about the paleopsych mailing list