Original title: face recognition is cracked? Alipay WeChat: in case of stolen brush, full payment.
Four seas network news recently, an American AI enterprise kneron (Chinese name Neng) claimed that one family could complete payment by making face masks, including WeChat payment, Alipay and railway face brushing. In this regard, experts said that the current face recognition technology does have a certain probability of being cracked by masks, suggesting that citizens protect their personal biological information.
The company claims to have done the experiment through a mask made by a Japanese company. This experiment has successfully broken through the verification of WeChat payment, Alipay brush face payment and train station face brushing. Take a look at the news. Knews found in the interview that most of the citizens are still in the wait-and-see stage.
In response, the Alipay team responded to previous attempts to contact the company to get detailed information, but the other side of the news and video off the shelf, did not provide more information. According to wechat, a number of technologies have been adopted, which can effectively resist attacks such as video, paper and mask. The two payment companies both said that if brush face payment leads to stolen brush, they can also apply for full compensation.
Experts say that with the current technology, there is still a certain probability that face recognition based on living body recognition will be cracked by masks. According to experts, it is possible to create 3D models and create high-precision masks by publishing your high-definition photos online. The best way is to protect your biological information.
The wheel of the era is rolling forward, science and technology are advancing constantly, and eyesight is everywhere. However, with the continuous development of technology, some of the disadvantages brought by science and technology have gradually emerged.
According to World Wide Web, one family of artificial Intelligent Company Kneron in Santiago has successfully deception many payment systems for face recognition, including Alipay and WeChat, with a special 3D mask. The team also claimed that they even entered China's railway stations in the same way.
Liu Juncheng, chief executive of kneron, said it showed that facial recognition technology did not meet security standards, which would pose a threat to user privacy.
In fact, it is not only this American company, but also China that has long had a case concerning the cracking of face wash payment.
In early August, a public official account for 3D printing released a video of the test. In the video, the staff used the wax figure made by 3D printing to deceive Alipay's face recognition system, and successfully bought a train ticket. The video was once crazy on the Internet and frightened many netizens, and it said that it had closed the function of face recognition.
In fact, this 3D printing head portrait is not original by Chinese people, and even less by American companies. Last year, a Japanese company had developed this product, which was very lifelike, and also attracted a lot of domestic media attention. At that time, many people said that face brushing and unlocking encountered new challenges.
More than that, other crack cases are also born at any time.
In October this year, the science team of class 402 of shangwai Xiuzhou foreign language school in Jiaxing revealed that they found in an extracurricular scientific experiment that they could replace a real person with a printed photo to swindle the Fengchao smart cabinet in the community and take out their parents' goods.
After that, Fengchao immediately went offline to "brush face and pick up pieces" and responded on the official microblog:
As for the feedback received recently, it is verified that the application is a beta version of trial operation, and it is being tested in a small scale. After receiving some user-friendly feedback, they have been offline for the first time. After improvement, you can pay attention to the Fengchao announcement. Thank you for your support and encouragement.
It is reported that the current face recognition technology can be divided into two categories: Based on 2D face image and based on 3D face image. When the camera of Fengchao express cabinet collects the face, it only forms a 2D plane image, which is equivalent to a photo.
2D face recognition only uses 2D camera to take plane imaging, so even if the algorithm and software are advanced, the security level is not high enough under limited information, and it is easy to crack with photos.
So, some people ask questions. Is WeChat and Alipay's face payment more common in life?
Wechat said: wechat face brush payment uses 3D living detection technology with the highest security level, and comprehensively uses 3D, infrared, RGB and other multimodal information, which can effectively resist attacks from video, paper, mask, etc.
At the same time, wechat payment will ensure account security through multi-dimensional security risk control strategy, and provide multi factor verification. Some users need to input the mobile phone number bound with wechat account or scan QR code for verification, which further improves the security.
The potential risks of account misuse and theft in wechat face swiping payment are extremely low probability events. If the account fund loss is caused by face swiping payment, you can also apply for full compensation.
Alipay changed its face in front of AI App'ZAO', triggering the risk of infringement, privacy and information security. The response said: Alipay's "brush face payment" adopts 3D face recognition technology. There are many kinds of face changing software, but no matter how realistic they are, they are unable to break the brush face payment.
On the one hand, although the parties insist that the security of technology is beyond doubt, on the other hand, Kneron claims to have deceived the system with a special 3D mask. Is WeChat and Alipay embarrassed?
At the 15th International Conference on information security and cryptography just held, Deng Huijie, Professor of network security at Singapore University of management, said that biometric technologies such as facial expressions and fingerprints have become increasingly prominent for personal information, data leakage and other network security issues, which makes people worried.
"Through experiments, we found that 70% of the total number of users can successfully unlock facial recognition in personal photos published on social platforms, that is to say, a photo you put on the Internet is likely to directly unlock your face recognition. "In this way, hackers can crack the biometric system through photos, videos and even sounds, and successfully steal user information," said Deng Huijie, Professor of network security at Anson, Singapore Management University.
He said that some face recognition systems are three-dimensional, which will make you nod and blink when authenticating, but it is not difficult for hackers. "They can unlock the 3D information of a person's face with the help of video, or cut out the eyes and mouth on the photos and replace them with usable ones. '
As the professor has just finished, the US artificial Intelligent Company has broken the face recognition technology of Alipay and WeChat. It's terrible to see how big the risk of face recognition is!
People are always afraid of unknown things, but they can not escape their temptation, because people are greedy. Because of greed, human beings are always prone to go to extremes, escape from reality, and do not want to give up some existing achievements.
Face recognition, as a big industry in artificial intelligence industry, has generated a wave of dividends in the industry, so it has been injected with too many bubbles.
Professor Wang Wenchao, a postdoctoral in biology from Harvard University and a senior scientist from medical college, once said that there is such a phenomenon in today's society. Both Internet giants and artificial intelligence unicorns are keen on running horses and encircling the land. Under the temptation of the big market cake, they are forced to make short profits. Therefore, the abuse of face recognition technology has caused some evils to the society Bad influence.
When it comes to the safety of biometrics, he expressed his concern that there are many advanced technologies in the world, which can be simply divided into external biometrics and internal biometrics. For example, common fingerprint recognition and face recognition belong to exobiological features. Because compared with other biometric technologies, their training data is the easiest to obtain and can be collected through social networking photos.
He said that exobiometric technology is easy to replicate and will change with age and other external effects. However, due to its low threshold, it has been touted by a large number of technology companies and widely used. However, the internal biometric identification technology with strong privacy, such as hand vein recognition and DNA, has been squeezed in the market share due to its high technical difficulty, which is a very bad situation Elephant.
Indeed, both government departments, social media and even start-ups have paid too much attention to face recognition technology, gathered too much resources, created a big bubble, and finally made others pay IQ taxes.
As a technology fanatic, I always believe that we should not waste too much energy on the research of technologies that are inherently more harmful than beneficial for the sake of short-term dividends.
I also hate those hypocrites who often say, 'every technology's emergence and development is initially well intentioned. Technology itself is not right. We can't stop eating because of choking. We have to give them a chance to try and error'.
I want to say that there is nothing wrong with technology, but does it make sense to make mistakes?