Can a robot think and feel? Can programming lead to something that imitates human intelligence and consciousness? Chinese artificial intelligence (AI) company Turing Robot works to build brains for robots, and envisions putting an AI robot in every household.
AI in the spotlight
The future when every household lives with an AI robot could be nearer than you thought.
In 1997, IBM’s chess-playing computer Deep Blue defeated world champion Gary Kasparov. Less than 20 years later, in March of 2016, Google’s deep learning computer program AlphaGO claimed victory over champion Lee Sedol at Go, a game which is widely considered to be the most complex board game in the world in terms of total possibilities and game permutations.
For there to exist AI that can compete with and even show signs of surpassing human intelligence, or that computers in general can exist – without which our world today would be quite different – we must pay tribute to the father of theoretical computer science and AI, Alan Turing.
“I respect Alan Turing because he was determined to make things happen, even when faced with great difficulties from all sides,” said Yu Zhichen, founder and CEO of Turing Robot.
Yu and his team have been working on the “upgraded AI”, something which was defined by Yu as the AI that actually makes a robot capable of reasoning and solving problems. The entrepreneur, still in his 30s said, “Before I retire at 60, I would devote my last 30 years to make the future happen, a future when upgraded AI comes true.”
Two major products: a brain and an OS for robots
Yu and his team have been working to integrate robotics with AI and internet. However, instead of making robots, they positioned themselves to build the brains of robots from the very beginning.
Founded in 2010, the company released the Turing Robot in 2014, an open platform for AI robotscovering the technologies of voice recognition, semantic recognition, and cognitive computing. It is a highly intelligent robot brain in the context of Chinese language recognition, with a semantic recognition rate reaching 94.7%.
In addition, it employs technologies including Deep QA and natural language understanding (NLU) to offer further services. As of this July, Turing Robot has accumulated more than 230,000 third-party developers, contributing to a total of 16.2 billion service requests.
Another major product is Turing OS, an AI-level robot operating system that provides natural and friendly modes for human-machine interaction. Released in late 2015, Turing OS is committed to helping their business partners’ efforts to make their own robots, or developing AI applications for their respective fields.
It consists of three engines: the emotion-computing engine that can recognize 25 human emotions via intonations and body language; the thinking-enhancement engine that trains robots to think like a human and greatly extends the duration of human-machine interaction; and the self-learning engine that allows robots to upgrade themselves with data collected from the interactions with users.
This late July witnessed the upgrade of the self-developed Turing OS 1.0 to Turing OS 1.5 at the company’s innovation conference. Some Turing robots with the latest system will meet the public in the second half of 2016.
Household robots hold the newest opportunities
Instead of straining to enter into the arena of developing industrial robots – the edge of which has been led by counterparts in countries like Germany and Japan, Turing Robot focuses on household robots.
Yu believes that the commercialization of AI and robotics is speeding up, and that human-machine cooperation may soon become popular in industries like education, healthcare, and finance. Nevertheless, for Turing Robot, the focus will stay on education and entertainment for children.
Among the 150 members of Turing Robot, about 83% are engineers from the world’s top-ranked universities, each with their own specialized area of expertise, each contributing to the company’s over 200 patents.
Last December, the company secured its Series B financing of RMB 100 million (USD 15 million), followed by another RMB 50 million from Alpha Group earlier this year. Yu said the funding has been used for R&D, service maintenance, and marketing.
The road ahead is long and… possibly bright
China’s Ministry of Industry and Information Technology, the National Development and Reform Commission and the Ministry of Finance jointly predicted that, by 2020, the market for service robots in China may reach a total of RMB 30 billion. Additionally, according to the International Federation of Robotics, the global market for service robots is estimated to reach USD 46.2 billion by 2017.
Though the main target at present is still within China, Yu said that Turing Robot will soon march into English-speaking countries, “Now is a great time, with an abundance of opportunities, and good products and services must go global for further improvements and more users.”
Turing Robot’s development may well serve Alan Turing’s vision, as he once said, “We can only see a short distance ahead, but we can see plenty there that needs to be done.”
(Top photo from Baidu Images)