SlideShare a Scribd company logo
Are you Human? (Sorry, I had to ask) Ecaterina Valică http://students.info.uaic.ro/~evalica/
Agenda What is CAPTCHA? Types of CAPTCHA Where to use CAPTCHAs? Guidelines when making a CAPTCHA Ways to break CAPTCHAs reCAPTCHA Human Computation Games
Example: Filling out a form Google uses CAPTCHA for Gmail accounts:
Beginnings C ompletely  A utomated  P ublic  T uring test to tell  C omputers and  H umans  A part  Created in 2000 for Yahoo to  prevent  automated e-mail account registration, by  Luis von Ahn , Manuel Blum, Nicholas Hopper and John Langford, Carnegie Mellon University.
Inventor  Luis von Ahn ( 1978 - ) Photograph by Mike McGregor
What is CAPTCHA? A  program  that can tell whether its user is a human or a computer. It uses a type of  challenge-response  test to determine that the response is not generated by a computer.
Turing Test „ Standard Interpretation"  player  C , the interrogator,  is tasked with trying to determine which player  -  A  or  B  - is a computer and which is a human.
Reverse Turing Test A CAPTCHA is sometimes described as a reverse Turing test, because it is  administered by a machine  and  targeted to a human .
So, CAPTCHA is… A program   that can generate and grade tests that: Most  humans   can  pass; Current  computer  programs  cannot  pass.
Making a CAPTCHA Pick  random string  of characters   (or words) ifhkfp Renders it into a  distorted  image
Making a CAPTCHA …  and the program generates a  test : Type the characters that appear in the image
Outperform the computers In many simple tasks, a typical 5-year-old can outperform the most powerful computers easier  for computers: like medical diagnosis, playing chess, hard  for computers:  operations requiring  vision ,  hearing ,  language  or  motor control .
Type: Early CAPTCHAs Generated by the EZ-Gimpy program; Used previously on Yahoo!
Type: Improved CAPTCHA  high contrast for human readability;  medium, per-character perturbation;  random fonts per character;  low  background noise ;
Type: A modern CAPTCHA rather than attempting to create a distorted background and high levels of warping on the text; focus on making segmentation difficult by adding an  angled line ;
Type: A modern CAPTCHA another way to make segmentation difficult is to  crowd symbols together ;  this can be read by humans but cannot be segmented by bots;
Other Types of CAPTCHA Animated CAPTCHAs 3D CAPTCHA ASCII art Reverse CAPTCHA  "Leave this field blank"
Other: Cognitive Puzzles Distinguish pictures of dogs from cats Choose a word that relates to all the images Trivia questions  Math and word problems  3D Object CAPTCHA Solve failed OCR inputs
Other: Distinguish pictures Microsoft  Asirra   (Animal Species Image Recognition for Restricting Access); KittenAuth  Project  .
Other :  Mathematical CAPTCHA
Other :  Mathematical CAPTCHA
Other: 3D Object CAPTCHA You must enter them in the  exact sequence  listed: The Head of the Walking Man, The Vase, The Back of the Chair.
Other: Jumble Game
Other:  Drupal  Examples
Other: Tests „ Common Sense" questions: „ What is 3 + 5?“ „ What color is the sky?" Type the word 'orange';  Require a valid email to approve; These attempts  violate  principles:  they cannot be automatically generated;  they can be easily cracked given the state of AI.
Where to use CAPTCHAs? Data Collection Worms and Spam Preventing Comment Spam in Blogs Protecting Email Addresses From Scrapers Online Polls Protecting Website Registration Preventing Dictionary Attacks Search Engine Bots
Where to use CAPTCHAs? Preventing Comment Spam in Blogs.   Protecting Email Addresses From Scrapers.   Mechanism to hide your email address, require users to solve a CAPTCHA before showing your email address Online Polls.  You cannot trust the results of an online roll because anybody could just write a program to vote for their favorite option thousands of times.
Where to use CAPTCHAs? Protecting Website Registration.  ( E-mail services: Yahoo, Microsoft, Google )   Preventing Dictionary Attacks   (in password systems). Prevent a computer to iterate through the entire space of passwords by requiring it to solve a CAPTCHA after a certain number of unsuccessful logins.  Search Engine Bots.  It is sometimes desirable to keep webpages unindexed to prevent others from finding them easily.
Guidelines  Image Security.  Images of text should be distorted randomly before being presented to the user. Script Level Security.  Insecurities:  Systems that pass the answer in plain text; Systems where a solution to the same CAPTCHA can be used multiple times ("replay attacks").
Guidelines Security Even After Wide-Spread Adoption.  There are CAPTCHAs that would be insecure if a significant number of sites started using them.  Example: text-based questions; A parser could easily be written that would allow bots to bypass the test; Such “CAPTCHAs” rely on the fact that  few  sites use them, and thus that a bot author  has no  incentive to program their bot to solve that challenge.
Guidelines Accessibility. CAPTCHAs prevent  visually impaired  users  (for example, due to a disability or because it is difficult to read)  from accessing the protected resource; They use screen reader, so when you reached an image, all it can do is to read the caption of that image; Solution : permitting users to opt for an  audio   or sound  CAPTCHA.
Guidelines:  Accesibility Hard to read CAPTCHAs:
Guidelines:  Accesibility Worst CAPTCHAs:
Ways to break CAPTCHAs Exploiting bugs in the implementation that allow the attacker to completely bypass the CAPTCHA; Improving Character Recognition software (OCR – Optical Character Recognition ); Using cheap human labor to process the tests ( sweatshops ).
Break:  Insecure implementation Re-using the  session ID  of a known CAPTCHA image.  Other CAPTCHA use a hash of the solution as a key passed to the client to validate. Often it is small enough  size  that it can be  cracked .  Other implementations use only a  small fixed pool  of CAPTCHA images (Asirra – 3 millions).
Break : Character Recognition Programs that have the following functions: Extraction of the image from the web page  Removal of background clutter, for example with color filters and detection of thin lines;  Segmentation , i.e. splitting the image into regions each containing a single letter;  Identifying the letter for each region.
Attacks –  EZ-Gimpy 2000 Yahoo's early CAPTCHA called "EZ-Gimpy“; The program picks a word from a dictionary, and produces a distorted and noisy image of the word; Algorithm  for breaking EZ-Gimpy ( 92% ): Locate possible letters at various locations;  Construct graph of consistent letters;  Look for plausible words in the graph.
Attacks –  EZ-Gimpy 2000 EZ-Gimpy Possible Letters Graph of Letters Plausible Words
Attacks –  SimpleOCR Engine 2002
Attacks – Jan/Feb 2008 Google ( Jan 17 ) 20% Hotmail (Feb 6) 30-35% Yahoo (Feb 22) 30-35%
Attacks – Projects Several broking CAPTCHAs projects: http://libcaca.zoy.org/wiki/PWNtcha http:// www.lafdc.com / captcha /
Break: Human solvers Attacks that uses humans to solve the puzzles; Approaches: relaying the puzzles to a group of human  operators  who can solve CAPTCHAs;  copying the CAPTCHA images and using them as CAPTCHAs for a  high-traffic site  owned by the attacker.
CAPTCHA Sweatshops A computer fills out a form and when it reaches a CAPTCHA, it gives it to the operator to solve.  Weakness for Asirra:  if the database of cat and dog photos can be downloaded, then paying workers $0.01 to classify each photo,  means that almost the entire database of photos can be deciphered for $30,000. Once IP has misclassified a challenge, a human needs to just solve two Asirras in a row from the same browser session.
CAPTCHA Sweatshops Not Economical Viable A typical spam run of 1 million messages per day would  cost $14,000  per day and require  116  people working 24/7.  $2.50 / h for each human 720 CAPTCHAs per hour per human 1/3 cent per account
Porn Companies (October 2007) They write a program that fills out the entire registration form (ex Yahoo); When the program gets to the CAPTCHA it can’t solve it; So it copies the CAPTCHA back to the porn page; One person gets the screen saying if you want to see the next picture, you’ve got to tell what word is in the specific CAPTCHA.
Porn Companies (October 2007)
Next CAPTCHA Generation CAPTCHAs can be made stronger, but they are already  too advanced  for a large percentage of Internet users;  CAPTCHA devolves from a simple human reading test into an intelligence test or an acuity test.
reCAPTCHA (2007) New form of CAPTCHA that also helps  digitize books ; The words displayed to the user come directly from old books that are being digitized;  Words that OCR  could not identify ;
reCAPTCHA Pairs an  unknown word  with a  known one ;  Distorts them both and puts a line through them and then sent them to be proofread; Respondent answers both elements:  half of effort validates the challenge;  the other half is captured as work.
reCAPTCHA
Time spent Roughly  60 million  CAPTCHAs are  solved  each day; Medium  10 seconds  to solve a captcha; People around the world waste more than  150,000 hours  on solving CAPTCHAs;
Time spent A fifth of those users giving  30,000 daily man-hours of work;  It would constitute the  world's fastest and most accurate character-recognition  computer, processing 10 million words a day.  Recreating the books – word by word
Time spent 9 Billion Human-Hours  of  Solitaire  were played in 2003 Empire State Building 7 million Human-Hours  (6.8 Hours of Solitaire) Panama Canal 20 Million Human-Hours  (Less than a day of Solitaire)
Wasted human cycles If the world's computer Solitaire players could be coaxed into  enjoying a game  that contributed to solving a computing problem, he calculates, it would produce billions of man-hours of labor each year.  „ make all of humanity more efficient by exploiting the human cycles that get wasted“
Wasted human cycles People will contribute their brainpower, but only if they're given an  enjoyable , time-killing experience in exchange.  Most projects that harness human processing power rely on a different motivator:  money .  Which produces better results —  a small group of experts or a huge mob of amateurs ?
Human Computation  Things that we humans can do and computer cannot, like: Labeling images with words; Picking out a voice in a loudly room; Humans have trouble remembering long, random strings of characters, yet they excel at remembering faces and objects.
Symbiotic relationship One in which humans solve some problems, computer solve some other problems; Image search  - A method that every image on the Web could give us accurate textual descriptions of those images;
The ESP Game Two-player online game; Partners don’t know each other  and can’t communicate; Object of the game:  Type the same word ;  The only thing in common is an image ;
The ESP Game Player 1 Guessing:  CAR Guessing :  HAT Guessing:  KID Success! You agree on CAR Player 2 Guessing :  BOY Guessing :  CAR Success! You agree on CAR
The ESP Game The ESP Game has been licensed (2006) by Google in the form of the  Google Image Labeler , and is used to improve the accuracy of the  Google Image Search .  “ 5000 people  playing simultaneously can  label all images on Google in  30 days !”
http://gwap.com/gwap/
http://gwap.com/gwap/ ESP  Tag a Tune   Matchin
http://gwap.com/gwap/ Squigl   Verbosity
Future Games Language translation.   A game could challenge two players who don’t speak the same language to translate text from one language to the other. Monitoring of security cameras .  Players could monitor security cameras and alert authorities about suspected illegal activity.
Future Games Improving Web search.   People have varying degrees of skill at searching for information on the Web. A game could be designed in which the players perform searches for other people. Text summarization.   Imagine a game in which people summarize important documents for the rest of the world.
Still not thinking big enough "If we have that many people all doing some little part, we could do something insanely huge for humanity."  "We'll never run out of things to digitize"
 
Bibliography Site:  Luis von  Ahn  Website  (2006) Site:  reCAPTCHA  (2007) Site:  CAPTCHA  (2007)  Site:  Gwap  (2008) Interview: „ Using  “ captchas ” to  digitize   books “ (2007) Interview:  „For  Certain   Tasks ,  the   Cortex  Still Beats  the  CPU“  (2007)
Bibliography Video: Wired –  „Human  Computation “  (2007) Video: Google TechTalks  –   “Human Computation”   (2006)  Paper: „ Games   With  a  Purpose “ (2006) Paper: „ How   Lazy   Cryptographers  do AI “  (2004) Paper: „ CAPTCHA:  Using  Hard AI Problems  for  Security “(2003)
Bibliography Article:  “CAPTCHA is Dead, Long Live CAPTCHA!”  (2008) Article:  „ Yahoo's CAPTCHA Security Reportedly Broken “  (2008) Article:  „ Anti-CAPTCHA   operations  on Microsoft Mail“  (2008) Article:  „ Google’s CAPTCHA busted in recent spammer tactics “  (2008)
Bibliography Paper:  „ Recognizing   Objects  in  Adversarial   Clutter “  (2002) Article:  Wikipedia  CAPTCHA  (2008) Article: „ CAPTCHA Effectiveness”  (2006)   Article:  „ Breaking  a Visual CAPTCHA“  (2002) Article:  „Human  or  Computer? Take  This  Test“  (2002) Site:  XKCD  (2008)
Thank you!

More Related Content

PPT
Captcha ppt
PPTX
PPTX
Captcha
PPTX
Captcha and Recaptcha Seminar
PPSX
Captcha
PPTX
CAPTCHA
PPT
captcha.ppt
PPT
Captcha1
Captcha ppt
Captcha
Captcha and Recaptcha Seminar
Captcha
CAPTCHA
captcha.ppt
Captcha1

What's hot (20)

PPTX
CAPTCHA
PPT
Captchas
PPTX
Introduction about Captcha and its types
PPTX
PPT
PPTX
Deepfakes: Trick or Treat?
PPTX
Computer vision and robotics
PPTX
Computer vision
PPTX
CAPTCHA
PPT
Face Detection techniques
PPT
CAPTCHA- Newly Attractive Presentation for Youth
PPTX
Computer Vision - Real Time Face Recognition using Open CV and Python
PPTX
Computer vision
PPTX
What is computer vision?
PDF
Captcha seminar report
PPTX
Computer vision
PDF
Computer vision basics
PDF
Visual Question Answering 2.0
PDF
[PR12] You Only Look Once (YOLO): Unified Real-Time Object Detection
PPTX
Computer vision ppt
CAPTCHA
Captchas
Introduction about Captcha and its types
Deepfakes: Trick or Treat?
Computer vision and robotics
Computer vision
CAPTCHA
Face Detection techniques
CAPTCHA- Newly Attractive Presentation for Youth
Computer Vision - Real Time Face Recognition using Open CV and Python
Computer vision
What is computer vision?
Captcha seminar report
Computer vision
Computer vision basics
Visual Question Answering 2.0
[PR12] You Only Look Once (YOLO): Unified Real-Time Object Detection
Computer vision ppt
Ad

Similar to Captcha (20)

PPTX
PPTX
CAPTCHA.pptx
PDF
Captcha a security measure against spam attacks
PPTX
PPTX
Captcha ppt
DOCX
PDF
Captcha Recognition and Robustness Measurement using Image Processing Techniques
PPTX
latest ppt in tranning
PDF
Evolution of captcha technologies
PDF
49 captcha
PDF
A CAPTCHA in the Rye
PPTX
Captcha seminar
PPTX
Captcha system
PPT
Jean captcha-ppt
PDF
14A81A05A8
PDF
A Survey of Current Research on CAPTCHA
DOCX
Captcha report
PPT
49 captcha
PDF
PDF
CAPTCHA.pptx
Captcha a security measure against spam attacks
Captcha ppt
Captcha Recognition and Robustness Measurement using Image Processing Techniques
latest ppt in tranning
Evolution of captcha technologies
49 captcha
A CAPTCHA in the Rye
Captcha seminar
Captcha system
Jean captcha-ppt
14A81A05A8
A Survey of Current Research on CAPTCHA
Captcha report
49 captcha
Ad

More from Ecaterina Moraru (Valica) (20)

PDF
UI/UX Tips & Tricks for developers - FOSDEM2020
PDF
UI/UX Tips & Tricks for developers
PDF
Sketching Session
PDF
CSS Grid vs. Flexbox
PDF
Designing in the open
PDF
XWiki Skin 10.x+ ideas
PDF
Difficulties in having more designers participate in Open Source
PDF
CSS Variables — Native reusable custom properties
PDF
Design proposals status 9.x
PDF
What's new in XWiki 8.x and half of 9.x
PDF
Expectations from Open Source Designers
PDF
Success stats from OSD community
PDF
Future of XWiki Skins
PDF
Design process in an Open Community
PDF
XWiki Usability Testing Sessions
PDF
About XWiki.org
PDF
Evolution of CSS
ODP
XWiki Improvements Review (ver 2.4 - 5.1)
PDF
Interconectarea Semantica A Datelor In Contextul Managementului Informatiilor...
UI/UX Tips & Tricks for developers - FOSDEM2020
UI/UX Tips & Tricks for developers
Sketching Session
CSS Grid vs. Flexbox
Designing in the open
XWiki Skin 10.x+ ideas
Difficulties in having more designers participate in Open Source
CSS Variables — Native reusable custom properties
Design proposals status 9.x
What's new in XWiki 8.x and half of 9.x
Expectations from Open Source Designers
Success stats from OSD community
Future of XWiki Skins
Design process in an Open Community
XWiki Usability Testing Sessions
About XWiki.org
Evolution of CSS
XWiki Improvements Review (ver 2.4 - 5.1)
Interconectarea Semantica A Datelor In Contextul Managementului Informatiilor...

Recently uploaded (20)

PPTX
Vacation Rental Market Scraping for Smarter Investment Strategies.pptx
PDF
Chardham Yatra Packing List 2026 – Essentials to Carry
PPTX
Copy of Japan Animated Template slidesppt.net.pptx
PPTX
What Should You Expect from a Colosseum Private Tour.pptx
PDF
Experience Divinity with Nashik Kumbh Mela Package 2027.pdf
PPTX
Quiz- Thursday.pptxaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
PDF
Shravan and Monsoon Travel Made Easy Stay Updated with Live Train Running Status
PDF
How Can Indians Visit Kailash Mansarovar.pdf
PDF
How to Choose the Best Tour Operators in Rajasthan – A Complete Guide.pdf
PPTX
MACRO-PERSPECTIVE-IN-HOSPITALITY-AND-TOURISM-MODULES.pptx
PPTX
Enjoy Peaceful Travel Even in Busy Places.pptx
PDF
Where is Kailash Mansarovar in India or China.pdf
PDF
How Expensive is Mansarovar Yatra cost from Mumbai.pdf
PDF
Explore Gujarat with the Best Tour Packages
PPTX
Gujarat Tour Packages – Spiritual, Cultural & Scenic Journeys
PDF
Golden Triangle Tour A Complete Travel Guide.pdf
PDF
Which Month is Best for Kailash Mansarovar Yatra.pdf
PDF
Explore Luxemburry.eu, the ancient of lands in Europe
PPTX
Minimalist Business Slides XL by Slidesgo.pptx
DOC
Best Astrologer in Agra .
Vacation Rental Market Scraping for Smarter Investment Strategies.pptx
Chardham Yatra Packing List 2026 – Essentials to Carry
Copy of Japan Animated Template slidesppt.net.pptx
What Should You Expect from a Colosseum Private Tour.pptx
Experience Divinity with Nashik Kumbh Mela Package 2027.pdf
Quiz- Thursday.pptxaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
Shravan and Monsoon Travel Made Easy Stay Updated with Live Train Running Status
How Can Indians Visit Kailash Mansarovar.pdf
How to Choose the Best Tour Operators in Rajasthan – A Complete Guide.pdf
MACRO-PERSPECTIVE-IN-HOSPITALITY-AND-TOURISM-MODULES.pptx
Enjoy Peaceful Travel Even in Busy Places.pptx
Where is Kailash Mansarovar in India or China.pdf
How Expensive is Mansarovar Yatra cost from Mumbai.pdf
Explore Gujarat with the Best Tour Packages
Gujarat Tour Packages – Spiritual, Cultural & Scenic Journeys
Golden Triangle Tour A Complete Travel Guide.pdf
Which Month is Best for Kailash Mansarovar Yatra.pdf
Explore Luxemburry.eu, the ancient of lands in Europe
Minimalist Business Slides XL by Slidesgo.pptx
Best Astrologer in Agra .

Captcha

  • 1. Are you Human? (Sorry, I had to ask) Ecaterina Valică http://students.info.uaic.ro/~evalica/
  • 2. Agenda What is CAPTCHA? Types of CAPTCHA Where to use CAPTCHAs? Guidelines when making a CAPTCHA Ways to break CAPTCHAs reCAPTCHA Human Computation Games
  • 3. Example: Filling out a form Google uses CAPTCHA for Gmail accounts:
  • 4. Beginnings C ompletely A utomated P ublic T uring test to tell C omputers and H umans A part Created in 2000 for Yahoo to prevent automated e-mail account registration, by Luis von Ahn , Manuel Blum, Nicholas Hopper and John Langford, Carnegie Mellon University.
  • 5. Inventor Luis von Ahn ( 1978 - ) Photograph by Mike McGregor
  • 6. What is CAPTCHA? A program that can tell whether its user is a human or a computer. It uses a type of challenge-response test to determine that the response is not generated by a computer.
  • 7. Turing Test „ Standard Interpretation" player C , the interrogator, is tasked with trying to determine which player - A or B - is a computer and which is a human.
  • 8. Reverse Turing Test A CAPTCHA is sometimes described as a reverse Turing test, because it is administered by a machine and targeted to a human .
  • 9. So, CAPTCHA is… A program that can generate and grade tests that: Most humans can pass; Current computer programs cannot pass.
  • 10. Making a CAPTCHA Pick random string of characters (or words) ifhkfp Renders it into a distorted image
  • 11. Making a CAPTCHA … and the program generates a test : Type the characters that appear in the image
  • 12. Outperform the computers In many simple tasks, a typical 5-year-old can outperform the most powerful computers easier for computers: like medical diagnosis, playing chess, hard for computers: operations requiring vision , hearing , language or motor control .
  • 13. Type: Early CAPTCHAs Generated by the EZ-Gimpy program; Used previously on Yahoo!
  • 14. Type: Improved CAPTCHA high contrast for human readability; medium, per-character perturbation; random fonts per character; low background noise ;
  • 15. Type: A modern CAPTCHA rather than attempting to create a distorted background and high levels of warping on the text; focus on making segmentation difficult by adding an angled line ;
  • 16. Type: A modern CAPTCHA another way to make segmentation difficult is to crowd symbols together ; this can be read by humans but cannot be segmented by bots;
  • 17. Other Types of CAPTCHA Animated CAPTCHAs 3D CAPTCHA ASCII art Reverse CAPTCHA "Leave this field blank"
  • 18. Other: Cognitive Puzzles Distinguish pictures of dogs from cats Choose a word that relates to all the images Trivia questions Math and word problems 3D Object CAPTCHA Solve failed OCR inputs
  • 19. Other: Distinguish pictures Microsoft Asirra (Animal Species Image Recognition for Restricting Access); KittenAuth Project .
  • 20. Other : Mathematical CAPTCHA
  • 21. Other : Mathematical CAPTCHA
  • 22. Other: 3D Object CAPTCHA You must enter them in the exact sequence listed: The Head of the Walking Man, The Vase, The Back of the Chair.
  • 24. Other: Drupal Examples
  • 25. Other: Tests „ Common Sense" questions: „ What is 3 + 5?“ „ What color is the sky?" Type the word 'orange'; Require a valid email to approve; These attempts violate principles: they cannot be automatically generated; they can be easily cracked given the state of AI.
  • 26. Where to use CAPTCHAs? Data Collection Worms and Spam Preventing Comment Spam in Blogs Protecting Email Addresses From Scrapers Online Polls Protecting Website Registration Preventing Dictionary Attacks Search Engine Bots
  • 27. Where to use CAPTCHAs? Preventing Comment Spam in Blogs. Protecting Email Addresses From Scrapers. Mechanism to hide your email address, require users to solve a CAPTCHA before showing your email address Online Polls. You cannot trust the results of an online roll because anybody could just write a program to vote for their favorite option thousands of times.
  • 28. Where to use CAPTCHAs? Protecting Website Registration. ( E-mail services: Yahoo, Microsoft, Google ) Preventing Dictionary Attacks (in password systems). Prevent a computer to iterate through the entire space of passwords by requiring it to solve a CAPTCHA after a certain number of unsuccessful logins. Search Engine Bots. It is sometimes desirable to keep webpages unindexed to prevent others from finding them easily.
  • 29. Guidelines Image Security. Images of text should be distorted randomly before being presented to the user. Script Level Security. Insecurities: Systems that pass the answer in plain text; Systems where a solution to the same CAPTCHA can be used multiple times ("replay attacks").
  • 30. Guidelines Security Even After Wide-Spread Adoption. There are CAPTCHAs that would be insecure if a significant number of sites started using them. Example: text-based questions; A parser could easily be written that would allow bots to bypass the test; Such “CAPTCHAs” rely on the fact that few sites use them, and thus that a bot author has no incentive to program their bot to solve that challenge.
  • 31. Guidelines Accessibility. CAPTCHAs prevent visually impaired users (for example, due to a disability or because it is difficult to read) from accessing the protected resource; They use screen reader, so when you reached an image, all it can do is to read the caption of that image; Solution : permitting users to opt for an audio or sound CAPTCHA.
  • 32. Guidelines: Accesibility Hard to read CAPTCHAs:
  • 33. Guidelines: Accesibility Worst CAPTCHAs:
  • 34. Ways to break CAPTCHAs Exploiting bugs in the implementation that allow the attacker to completely bypass the CAPTCHA; Improving Character Recognition software (OCR – Optical Character Recognition ); Using cheap human labor to process the tests ( sweatshops ).
  • 35. Break: Insecure implementation Re-using the session ID of a known CAPTCHA image. Other CAPTCHA use a hash of the solution as a key passed to the client to validate. Often it is small enough size that it can be cracked . Other implementations use only a small fixed pool of CAPTCHA images (Asirra – 3 millions).
  • 36. Break : Character Recognition Programs that have the following functions: Extraction of the image from the web page Removal of background clutter, for example with color filters and detection of thin lines; Segmentation , i.e. splitting the image into regions each containing a single letter; Identifying the letter for each region.
  • 37. Attacks – EZ-Gimpy 2000 Yahoo's early CAPTCHA called "EZ-Gimpy“; The program picks a word from a dictionary, and produces a distorted and noisy image of the word; Algorithm for breaking EZ-Gimpy ( 92% ): Locate possible letters at various locations; Construct graph of consistent letters; Look for plausible words in the graph.
  • 38. Attacks – EZ-Gimpy 2000 EZ-Gimpy Possible Letters Graph of Letters Plausible Words
  • 39. Attacks – SimpleOCR Engine 2002
  • 40. Attacks – Jan/Feb 2008 Google ( Jan 17 ) 20% Hotmail (Feb 6) 30-35% Yahoo (Feb 22) 30-35%
  • 41. Attacks – Projects Several broking CAPTCHAs projects: http://libcaca.zoy.org/wiki/PWNtcha http:// www.lafdc.com / captcha /
  • 42. Break: Human solvers Attacks that uses humans to solve the puzzles; Approaches: relaying the puzzles to a group of human operators who can solve CAPTCHAs; copying the CAPTCHA images and using them as CAPTCHAs for a high-traffic site owned by the attacker.
  • 43. CAPTCHA Sweatshops A computer fills out a form and when it reaches a CAPTCHA, it gives it to the operator to solve. Weakness for Asirra: if the database of cat and dog photos can be downloaded, then paying workers $0.01 to classify each photo, means that almost the entire database of photos can be deciphered for $30,000. Once IP has misclassified a challenge, a human needs to just solve two Asirras in a row from the same browser session.
  • 44. CAPTCHA Sweatshops Not Economical Viable A typical spam run of 1 million messages per day would cost $14,000 per day and require 116 people working 24/7. $2.50 / h for each human 720 CAPTCHAs per hour per human 1/3 cent per account
  • 45. Porn Companies (October 2007) They write a program that fills out the entire registration form (ex Yahoo); When the program gets to the CAPTCHA it can’t solve it; So it copies the CAPTCHA back to the porn page; One person gets the screen saying if you want to see the next picture, you’ve got to tell what word is in the specific CAPTCHA.
  • 47. Next CAPTCHA Generation CAPTCHAs can be made stronger, but they are already too advanced for a large percentage of Internet users; CAPTCHA devolves from a simple human reading test into an intelligence test or an acuity test.
  • 48. reCAPTCHA (2007) New form of CAPTCHA that also helps digitize books ; The words displayed to the user come directly from old books that are being digitized; Words that OCR could not identify ;
  • 49. reCAPTCHA Pairs an unknown word with a known one ; Distorts them both and puts a line through them and then sent them to be proofread; Respondent answers both elements: half of effort validates the challenge; the other half is captured as work.
  • 51. Time spent Roughly 60 million CAPTCHAs are solved each day; Medium 10 seconds to solve a captcha; People around the world waste more than 150,000 hours on solving CAPTCHAs;
  • 52. Time spent A fifth of those users giving 30,000 daily man-hours of work; It would constitute the world's fastest and most accurate character-recognition computer, processing 10 million words a day. Recreating the books – word by word
  • 53. Time spent 9 Billion Human-Hours of Solitaire were played in 2003 Empire State Building 7 million Human-Hours (6.8 Hours of Solitaire) Panama Canal 20 Million Human-Hours (Less than a day of Solitaire)
  • 54. Wasted human cycles If the world's computer Solitaire players could be coaxed into enjoying a game that contributed to solving a computing problem, he calculates, it would produce billions of man-hours of labor each year. „ make all of humanity more efficient by exploiting the human cycles that get wasted“
  • 55. Wasted human cycles People will contribute their brainpower, but only if they're given an enjoyable , time-killing experience in exchange. Most projects that harness human processing power rely on a different motivator: money . Which produces better results — a small group of experts or a huge mob of amateurs ?
  • 56. Human Computation Things that we humans can do and computer cannot, like: Labeling images with words; Picking out a voice in a loudly room; Humans have trouble remembering long, random strings of characters, yet they excel at remembering faces and objects.
  • 57. Symbiotic relationship One in which humans solve some problems, computer solve some other problems; Image search - A method that every image on the Web could give us accurate textual descriptions of those images;
  • 58. The ESP Game Two-player online game; Partners don’t know each other and can’t communicate; Object of the game: Type the same word ; The only thing in common is an image ;
  • 59. The ESP Game Player 1 Guessing: CAR Guessing : HAT Guessing: KID Success! You agree on CAR Player 2 Guessing : BOY Guessing : CAR Success! You agree on CAR
  • 60. The ESP Game The ESP Game has been licensed (2006) by Google in the form of the Google Image Labeler , and is used to improve the accuracy of the Google Image Search . “ 5000 people playing simultaneously can label all images on Google in 30 days !”
  • 64. Future Games Language translation. A game could challenge two players who don’t speak the same language to translate text from one language to the other. Monitoring of security cameras . Players could monitor security cameras and alert authorities about suspected illegal activity.
  • 65. Future Games Improving Web search. People have varying degrees of skill at searching for information on the Web. A game could be designed in which the players perform searches for other people. Text summarization. Imagine a game in which people summarize important documents for the rest of the world.
  • 66. Still not thinking big enough "If we have that many people all doing some little part, we could do something insanely huge for humanity." "We'll never run out of things to digitize"
  • 67.  
  • 68. Bibliography Site: Luis von Ahn Website (2006) Site: reCAPTCHA (2007) Site: CAPTCHA (2007) Site: Gwap (2008) Interview: „ Using “ captchas ” to digitize books “ (2007) Interview: „For Certain Tasks , the Cortex Still Beats the CPU“ (2007)
  • 69. Bibliography Video: Wired – „Human Computation “ (2007) Video: Google TechTalks – “Human Computation” (2006) Paper: „ Games With a Purpose “ (2006) Paper: „ How Lazy Cryptographers do AI “ (2004) Paper: „ CAPTCHA: Using Hard AI Problems for Security “(2003)
  • 70. Bibliography Article: “CAPTCHA is Dead, Long Live CAPTCHA!” (2008) Article: „ Yahoo's CAPTCHA Security Reportedly Broken “ (2008) Article: „ Anti-CAPTCHA operations on Microsoft Mail“ (2008) Article: „ Google’s CAPTCHA busted in recent spammer tactics “ (2008)
  • 71. Bibliography Paper: „ Recognizing Objects in Adversarial Clutter “ (2002) Article: Wikipedia CAPTCHA (2008) Article: „ CAPTCHA Effectiveness” (2006) Article: „ Breaking a Visual CAPTCHA“ (2002) Article: „Human or Computer? Take This Test“ (2002) Site: XKCD (2008)