The article discusses CAPTCHAs, which are tests designed to distinguish humans from computers. CAPTCHAs are based on the idea that humans can pass certain tests that current computer programs cannot. The term CAPTCHA stands for "Completely Automated Public Turing Test to Tell Computers and Humans Apart." The P in CAPTCHA stands for Public, meaning the code and data used by a CAPTCHA should be publicly available. Examples of CAPTCHAs include Gimpy, which asks users to identify three words from a distorted image; Bongo, which asks users to recognize a pattern in two series of blocks; and Pix, which asks users to identify images of objects. However, some CAPTCHAs, like Pix, may not be secure if the image database is not publicly available. CAPTCHAs have various applications, including preventing bots from creating multiple email accounts, preventing search engine bots from indexing web pages, and preventing shopping agents from accessing prices without seeing advertisements. However, CAPTCHAs can be bypassed by "stealing cycles from humans," where bots use humans to solve CAPTCHA tests. The authors argue that CAPTCHAs are useful because they help solve hard AI problems. They believe that if a program can pass a CAPTCHA test, it can be used to solve a hard AI problem. This is because solving a CAPTCHA test is as hard as solving a hard AI problem. The article also acknowledges the work of others who have developed similar systems, such as Andrei Broder's system at Altavista. The authors also mention the Open Mind Project, which uses people on the internet to help teach computers simple concepts. In conclusion, CAPTCHAs are a useful tool for distinguishing humans from computers and have various applications in the online world. They are based on the idea that humans can pass certain tests that current computer programs cannot, and they are designed to be publicly available.The article discusses CAPTCHAs, which are tests designed to distinguish humans from computers. CAPTCHAs are based on the idea that humans can pass certain tests that current computer programs cannot. The term CAPTCHA stands for "Completely Automated Public Turing Test to Tell Computers and Humans Apart." The P in CAPTCHA stands for Public, meaning the code and data used by a CAPTCHA should be publicly available. Examples of CAPTCHAs include Gimpy, which asks users to identify three words from a distorted image; Bongo, which asks users to recognize a pattern in two series of blocks; and Pix, which asks users to identify images of objects. However, some CAPTCHAs, like Pix, may not be secure if the image database is not publicly available. CAPTCHAs have various applications, including preventing bots from creating multiple email accounts, preventing search engine bots from indexing web pages, and preventing shopping agents from accessing prices without seeing advertisements. However, CAPTCHAs can be bypassed by "stealing cycles from humans," where bots use humans to solve CAPTCHA tests. The authors argue that CAPTCHAs are useful because they help solve hard AI problems. They believe that if a program can pass a CAPTCHA test, it can be used to solve a hard AI problem. This is because solving a CAPTCHA test is as hard as solving a hard AI problem. The article also acknowledges the work of others who have developed similar systems, such as Andrei Broder's system at Altavista. The authors also mention the Open Mind Project, which uses people on the internet to help teach computers simple concepts. In conclusion, CAPTCHAs are a useful tool for distinguishing humans from computers and have various applications in the online world. They are based on the idea that humans can pass certain tests that current computer programs cannot, and they are designed to be publicly available.