At SteerPoint, we’re always on the cutting edge of tech updates, so imagine how our interest was piqued when we discovered a computer program called Eugene Goostman passed the Turing test at a competition in England. Or, he supposedly passed it, depending on who you’re talking to.
For the unfamiliar, the Turing test supposedly measures whether or not a machine is capable of thinking. The test has been around for 65 years, and to date, no computer program has ever passed it. The metric for the Turing test is that a machine has to trick judges into thinking it is human at least 30% of the time. The Eugene Gootsman chatbox managed a score of 33%.
Of course, there were a few “advantages” that the Eugene Gootsman chatbox had built-in: first, it pretended it was a 13-year-old boy from the Ukraine, so English wasn’t his first language, and some have accused the chatbox of deflecting questions rather than answering them.
Naysayers point out that the Turing test isn’t exactly what most people believe it is. The previous link is the literal paper where Alan Turing describes the test (and we don’t blame you if you don’t want to read the whole thing), but in the paper he implies the Turing test wasn’t meant to be an actual indicator of intelligence, but more of Turing’s predictions on where technology would be in the future.
According to that interpretation of this past weekend’s computer technology, we haven’t stumbled upon a new age of artificial intelligence, but rather finally caught up to Turing’s idea of where technology would be at this point.
Additionally, many modern day experts believe the Turing test isn’t really the best measurement of artificial development anyhow, and that many more impressive accomplishments have been done with artificial intelligence rather than tricking humans into conversation.
If the idea of artificial intelligence sounds frightening to you, it’s worth noting that we interact with intelligent programming all the time. Think about the number of people who get information from Siri, Apple’s virtual assistant. Google is in the process of developing a self-driving car. Have you ever used Netflix and taken the program’s suggestions on what to watch next? It’s likely you find all of this far more useful and impressive than a chatbox being able to trick humans into conversing 30% of the time.
Whether or not you believe that passing the Turing test is impressive, it still seems like we won’t be seeing Space Odyssey visions of HAL anytime soon.