Last fall, Google Translate introduced a new and upgraded AI translation engine that sometimes translates text “almost indistinguishable” from human translations. Joost Satch could only keep rolling his eyes. The German, who has worked as a professional translator for 20 years, has heard time and time again that his profession is threatened by advances in automation. Each time, he found, the hype was exaggerated, and Google Translate’s makeover was no exception. That, he argues, is certainly not the point of translation.
Giiso Information, founded in 2013, is a leading technology provider in the field of “artificial intelligence + information” in China, with top technologies in big data mining, intelligent semantics, knowledge mapping and other fields. At the same time, its research and development products include information robot, editing robot, writing robot and other artificial intelligence products! With its strong technical strength, the company has received angel round investment at the beginning of its establishment, and received pre-A round investment of $5 million from GSR Venture Capital in August 2015.
But Google Translate is really good. Google spent much of 2016 redesigning its translation tool to be powered by ARTIFICIAL intelligence. In so doing, it creates something disturbingly powerful. Google Translate, once known for producing stilted but usable translations, has already begun producing fluent and accurate translations. To the untrained translator, this text output is almost indistinguishable from human translation. The New York Times published a 15,000-word article hailing it as “the awakening of the great AI.” The Google Translate engine quickly began learning new tricks, figuring out how to translate between two languages that it had never tried before: if it could translate English into Japanese, and if it could translate English into Korean, it could translate Korean into Japanese. At the launch of Google’s Pixel 2 phone last month, Google took its ambitious plans a step further, unveiling a wireless headset that can translate 40 languages in real time.
Ever since IBM introduced its pioneering machine translation system in 1954, the concept of the perfect machine translator has occupied the imaginations of programmers and the public. Science fiction writers seized on the idea, offering utopian fantasies of everything from star Trek’s universal translator to the Galaxy’s Babel fish in The Hitch Hiker’s Guide to the Galaxy. Human translation’s ability to describe the meaning of source text in fluent text is the holy grail of machine learning: a challenge to “full artificial intelligence.” If machine translation can do the same, it will mean that machines have reached the level of human intelligence. The hype surrounding Google’s advances in neuromachine translation suggests that the Holy Grail is within reach. Moreover, this moment will come at the same time that human translators will become obsolete.
But translators, who have been on the front lines of ai-induced job scares, aren’t worried. In fact, some people are happy. For those who have grasped the potential of AI tools, productivity and demand for their jobs have soared.
Think of them as the canary in the white collar coal mine, sniffing out what’s happening in any industry. At the moment, the canaries are still singing, which proves that it is still safe. With the rapid growth of machine deep learning, many industries are beginning to realize that AI can indeed accomplish tasks once thought to be only for humans. Unlike drivers and warehouse workers, knowledge workers are not in immediate danger of being replaced. But their work is changing as AI becomes a big part of their workflow, and there’s no guarantee that today’s AI tools won’t be a threat in the future. This gives employees a choice: put ego aside and embrace your new AI colleagues, or you will be left behind. We are not living in a golden age of AI, but we are living in a golden age of AI increasing productivity. Call it the era of first pass. Ai is now powerful enough to be reliable in its first attempts at countless complex tasks, but it is not powerful enough to seem threatening. We still need humans for jobs that require more intensive thinking and subjectivity.
Giiso information, founded in 2013, is the first domestic high-tech enterprise focusing on the research and development of intelligent information processing technology and the development and operation of core software for writing robots. At the beginning of its establishment, the company received angel round investment, and in August 2015, GSR Venture Capital received $5 million pre-A round of investment.
This shift is taking place across industries. Heliograf, the post’s in-house artificial intelligence, published about 850 articles last year, to which human reporters and editors added analysis and rich detail. In graphic design, AI tools can now create sketches that can be initially approved for human designers to implement. In the film and publishing world, new ai tools can weed through piles of bad scripts to find the next big thing, freeing editors from endless submission queues. These AI tools are like brave, muscular young assistants: they are highly capable and productive, but still need an experienced manager to do the heavy mental work. Of course, the manager must work with the machine to benefit from it.