Gmail Audio CAPTCHA Broken (Tony Miller )


Subject: Gmail Audio CAPTCHA Broken
From:    Tony Miller  <antonio.miller@xxxxxxxx>
Date:    Fri, 2 May 2008 17:10:28 -0400
List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>

------=_Part_11079_18113903.1209762628930 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline A slashdot posting today got me thinking about auditory tasks which require a cognitive component which computers do not (yet) possess. http://tech.slashdot.org/article.pl?sid=08/05/02/1426253 The Microsoft Research Asirra "cats&dogs" CAPTCHA seems like an elegant visual solution to the problem -- low cognitive load on the user, high 'cognitive' load on the computer. I'm wondering if there is an auditory categorization task that is analogous to the Asirra scheme? -Tony P.S. If you go to the gmail signup page now, it sounds like Google has decreased the SNR of the auditory task in an effort to thwart this type of attack. The resulting audio is pretty annoying. ------=_Part_11079_18113903.1209762628930 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline A slashdot posting today got me thinking about auditory tasks which require a cognitive component which computers do not (yet) possess.<br> <br> <a href="http://tech.slashdot.org/article.pl?sid=08/05/02/1426253" target="_blank">http://tech.slashdot.org/article.pl?sid=08/05/02/1426253</a><br> <br> The Microsoft Research Asirra &quot;cats&amp;dogs&quot; CAPTCHA seems like an elegant visual solution to the problem -- low cognitive load on the user, high &#39;cognitive&#39; load on the computer. &nbsp;I&#39;m wondering if there is an auditory categorization task that is analogous to the Asirra scheme?<br> <br> -Tony<br><br>P.S. If you go to the gmail signup page now, it sounds like Google has decreased the SNR of the auditory task in an effort to thwart this type of attack.&nbsp; The resulting audio is pretty annoying.<br> ------=_Part_11079_18113903.1209762628930--


This message came from the mail archive
http://www.auditory.org/postings/2008/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University