kitten-on-gaurd.jpgKevin Larson, a researcher at Microsoft Corp.’s Advanced Reading Technologies Group, has come up with a novel way of defeating SPAM emails—asking the sender to identify a photo of a kitten. This is because computers that use automated programs to sign up for (then begin sending SPAM from) email accounts cannot identify the subject of a picture. The photograph can thus be used to authenticate that there is indeed a real person at the other end of the tube series.

This concept is not new (although the kitten, by default, must be) and commonly uses Human Interactive Proofs to verify that humanity is indeed still in control of the machine. HIPs, as they are more commonly (and conveniently!) referred to, require a human user to identify a series of random letters within a jumbled image. Computer programs have traditionally had trouble identifying these letters out of the pattern, one that the human eye can decipher at a glance.

Unfortunately, Spammers have been developing more and more advanced programs to beat the HIPs. So email as well as security providers are now racing against the clock to discover a way to beat the decoding systems and hold back the virtually limitless wall of SPAM another day. The problem with the picture of the kitten (other than some people are dog lovers) is that email or security providers would have to create a mammoth library of pictures to choose from, else risk the chance that of their opponents mapping the entire cache, thus rendering the technique useless.

Solution suggestions have ranged from a constantly updated series of pictures of a kitten jumping around a room to calligraphy set upon ornate backgrounds designed to fool a computer’s automated decoding program. Thus far, there’s been moderate success with HIPs and the development of image-based successors. So if you’re concerned with email security and anti-spam developments, just watch for fuzzy kittens.