In Sunday, IEEE SMC Society (IEEE System, Man, and Cybernetics Society) and the Chinese Institute of Automation, the Institute of Automation, CAS held an international symposium on Artificial Intelligence and cybernetics in Beijing.
During the Conference, experts from the field of artificial intelligence from several countries received an interview with the media, including the titanium media, to explore the challenges and opportunities that artificial intelligence may encounter in the future development process. Recently, it has been reported that some voice control assistants may be controlled by ultrasound and pose a security threat.
On this issue, Adrian Stoice, a senior researcher at NASA's Jet Propulsion Laboratory, sees that any technology can be exploited by bad guys, not just voice assistants, but any technology.
He illustrated, for example, that some toys have cameras, hackers can see the situation inside their home, if you have a robot, someone will connect it to use, then other people can see your face situation, revealing personal privacy. But this is a very small aspect, and it is impossible to prevent us from continuing to develop technology, we will have some protection technology to protect the network security, preventing hacker behavior.
' About the relationship between artificial intelligence and block chain, from the Institute of Automation of the Chinese Academy of Sciences, the director of the State Key laboratory of complex Systems management and control Wang Fei said that the large data itself does not have centralization tendency, the structure of data is often unstructured and fragmented.
Large data and block chains are all centered, or distributed, but going to the center itself means the whole center, so the data itself does not have the nature of centralization. ' Because big Data integrates all very small data, it's called big, and if there's no one who dares to use it, it's like building a house, if the floor is not high on the ground or sand foundation, the lid should be built on the reinforced concrete.
'
In addition to the block chain this technology, it is equivalent to reinforced concrete, you can put large data or artificial intelligence technology cover on the reinforced concrete, from this point of view he supports the study of block chain. Artificial intelligence is a small fire, developed 60 years, can only solve the problem of toys, basically in addition to some data in recent years to solve some of the larger problems, such as image recognition, speech recognition, natural language processing problems, but still have a lot of problems can not solve, why? Because there is no reinforced concrete foundation, Only small fires.
' Over the years, we have likened artificial intelligence to alchemy, a small fire can be alchemy, but to make the industry will be a raging blaze. Wang Leap feel that the block chain in the future can provide artificial intelligence raging fire, so these two are closely related, this is the block chain intelligence.
Intelligence can not be built on the soil base, on the Shaki, to build on the reinforced concrete, I believe this will lead to the whole social industrial management paradigm shift. Speaking of the market for consumers of all kinds of educational robots, in his view, and the robot has no relationship, such as the so-called Service robot is also pendulum window-dressing, the development of the robot itself has many important areas, such as the production line of industrial robots.
Relatively speaking, the recent domestic and foreign giants in pursuit of smart speakers is a natural pure human-computer interaction, the King leap Affirmation. For the smart phone this year, the hype of AI chip, Wang Leap said no such statement. ' I don't think there is a so-called AI chip now, of course it can be advertised.
'
Andreas Nuernberger, a professor at Magdeburg University in Germany, also agrees with the idea that it is not known who is pushing behind, thus becoming a hot word. ' Now the development of the chip accelerates the process of deep learning, in fact, before the image of this depth of learning. You now call them AI chips, but this is a product that is produced for a specific purpose, and I agree that there is no AI chip.
' For unmanned driving, Cao Dongpu, a professor at the University of Waterloo in Canada, said it might take a long time to land because the situation in different countries is not the same.
On the other side, the United States has a faster landing than China and Europe, mainly because their traffic scenes are simpler and the roads are much better. ' China is the common unmanned vehicle, for example, if the unmanned cars want to be in the city completely autonomous, I do not estimate ten or twenty years, (landing) can choke.
' In the short term, if the landing, basically is to limit the scene, the more limited the more the landing will be faster. ' For example, a set of scenes, mines, ports, as well as short parking logistics and sweeping cars, these engineering and municipal vehicles, with special conditions, in a year or two or two or three years, landed.