Brain wave-scanning helmet developed by scientists to help Chinese censors better detect porn

Chinese scientists have reportedly developed and tested a device that aims to help online censors better police pornography in the country.

Researchers at Beijing Jiaotong University in China created a helmet that can track the brain waves of its wearers. The scientists published their findings in the domestic peer-reviewed Journal of Electronic Measurement and Instrumentation earlier this month.

In China, where watching porn is illegal, the government employs mostly women as so-called porn appraisers, or “jian huang shi,” who screen online photo and video content. These human censors make use of artificial intelligence (A.I.) to help them in flagging explicit content faster.

Such a technique, however, is not foolproof, as some restricted videos and images still manage to slip past both the human censors and the A.I. algorithms.

According to the researchers, A.I.-based algorithms face challenges in recognizing images when they contain complex backgrounds.

While humans are far superior to A.I. in image recognition, they note that they too can be prone to errors when tired or after long periods of task repetition.

In their experiment, the research team tasked a group of 15 male volunteers between the ages of 20-25 years to look through a series of content that included explicit images.

During the testing phase, the researchers had to cover the sensitive parts of the images they displayed to comply with the law against pornographic materials.

The scientists found that observers raised their eyebrows when shown a nude image, even if it was shown for half a second amongst a stream of other images. The helmet can reportedly detect any spike in brain waves resulting from explicit content.

Porn appraisers with the helmet only need to face a computer screen for the device to automatically screen a large number of photos displayed and flag pornography.

As the device is able to adapt to the brainwaves of its user, it can filter out noises in the brain signal which may be caused by emotion, weariness or other distractions.

The researchers admit that the device is far from perfect. While it was able to detect almost all of the pornographic content, it also raised some false alarms.

The team placed the current overall accuracy of the device at 80 percent, which it attributes to a lack of available training materials.

Earlier this year, China’s cybersecurity regulator, the Cyberspace Administration of China,launched a “purification” campaign to “clean up” Chinese social media users’ “illegal” online content.

In addition to scrubbing pornography, the initiative to purify the internet also includes crackdowns on content that involve cyberbullying, online fraud, suicide and violence.

Featured Image via ulrichw

Share this Article