CAPTCHA Solving Economy

I was perusing my RSS feeds this morning and came across an article from ZDNET entitled, Inside India’s CAPTCHA Solving Economy. A CAPTCHA is a type of challenge-response test used in computing to ensure that the response is not generated by a computer (wikipedia).

I had never thought about the potential economics about setting up a service to “crack/break” CAPTCHA’s but a few enterprising individuals in India apparently have.

No CAPTCHA can survive a human that’s receiving financial incentives for solving it, and with an army of low-waged human CAPTCHA solvers officially in the business of “data processing” while earning a mere $2 for solving a thousand CAPTCHA’s, I’m already starting to see evidence of consolidation between India’s major CAPTCHA solving companies. The consolidation logically leading to increased bargaining power, is resulting in an international franchising model recruiting data processing workers empowered with do-it-yourself CAPTCHA syndication web based kits, API keys, and thousands of proxies to make their work easier, and the process more efficient.

I like how they call this Data Processing. I guess it really is. This reminds me of Gold Farming that generally takes place in games like World of Warcraft.. that’s data processing too I guess.

Looks like CAPTCHA might be on the decline if marketers start buying more from these solving services, so what’s next?

If you like what you read, follow me on

http://marketingtechblog.com Douglas Karr

I actually love seeing CAPTCHA’s demise. The technology is one of those irritating “Let’s interfere with everyone’s usability and experience because there are a couple of bad seeds out there.” The CAPTCHA technology needs to be challenged for a better technology that does not interrupt the typical user.
http://blog.botfu.com Kevin Marshall

I’m for seeing the captcha go away as well…in fact, http://www.semantichacker.com ran a contest earlier this year, and my entry specifically involved the idea of a semantic filter that could/should be used for this same type of thing that CAPTCHA is intended for…the basic idea being that every content item could/should have a semantic score that can be used to determine if a comment or other bit of content is relevant enough to be public/posted…from my point of view, automated responses are fine so long as they are relevant (think adsense vs spam)…a semantic filter would not stop spam, but it would force them to mold their messages into a more relevant message to the system they spam…and the hope is that, since it’s more relevant, it wouldn’t be quite so annoying to come across…

btw – no word yet on if they liked the prototype or idea…the notice is supposed to be sent out in the next week or two for results of the contest, so I guess we’ll see.

Stay Connected

Recent Comments

Subscribe to Rss Feed