What Can AI Do to Combat Piracy?

Use machine learning, deep learning, big data, cloud computing and other technologies to achieve copyright monitoring, registration and protection, and combat piracy.

Today, an article by the CEO of Sifou put CSDN at the center of public opinion.

The reason is that some authors of Think Tank found that their paid courses were pirated on the download channel of CSDN. As a well-known community communication platform, the CSDN download channel occupies 30 % of traffic, but it is not very effective in content supervision.

In the article, the author points out that the infringement and piracy incidents on CSDN are by no means isolated cases, and there seems to be no effective way to curb such infringement and piracy incidents on the Internet.

Similar content piracy is also common in the music and film and television industries. For example, the sci-fi blockbuster "The Wandering Earth" released during the Spring Festival this year was exposed as a pirated resource on the second day of its release, and was sold online for a few yuan. And it was not the usual "pirated" blurry "gun version", most of which were high-definition sources. It is really rampant.

Fighting Piracy****What's the difficulty?

The reason why pirated and counterfeit products are rampant is that these products are often shared for free or at low prices, and in the Internet age, their spread is even more difficult to curb.

In addition to the complex network environment, it is often difficult to find, there is another reasonInadequate punishment.

According to current relevant regulations, even if the film source is leaked, the compensation amount is only twice the original contract price, or even lower; in addition, although there are administrative penalties for infringements such as piracy, there is a lack of compulsory measures and the deterrent effect is insufficient.

If this matter had not been discovered and exposed by multiple authors, perhaps many people would still be kept in the dark.

The following figures show how big a market piracy is. A 2017 study by Frontier Economics suggests that by 2022, the total value of counterfeit and pirated goods could rise to**$2.81 trillion**(RMB 19.11 trillion. China accounts for 1.3% of the world’s total counterfeit and pirated goods.**46%**The study also estimates that net job losses related to counterfeiting and piracy could be as high as 5.4 million by 2022.

Faced with such a rampant piracy industry, it seems necessary to use artificial intelligence to combat it.

Automated copyright protection

Red Points, a Barcelona-based IP and brand protection startup, is using technology to solve this global problem.

The company's SaaS-based cloud solution system searches for counterfeit products or pirated content on the Internet on behalf of customers and automatically removes them, thereby providing enterprises with online IP infringement detection and automated removal of counterfeit information and pirated content.

The company's CEO Urquizu believes that machine learning tools, such asComputer Vision and Natural Language Processing,It can help brands search for information and detect counterfeits faster, therebyAutomatically enforce intellectual property protection.

“As opposed to service-based approaches, only scalable technology solutions can significantly address the challenges raised by IP infringement,” Urquizu said.

The company's anti-piracy approach is broken down into the following steps:

**track:**Online IP infringements are detected and eliminated through keyword monitoring systems that are managed by customers through a simple platform. In addition, copyright protection experts identify and update pirated sources every day, and machine learning uses detection records to add new search rules and continuously improves itself.

**delete:**Automatically enforce copyright and follow up with sources. Once violations are discovered, use automated rules to immediately send takedown and contact indexing requests, and take out-of-court action to stop further infringement.

**Report:**After discovering infringement, customers can receive customized reports to check the scope of their intellectual property protection. The platform will also identify other infringement trends and track the source of the infringer.

In addition, by using**Machine learning and deep learning techniques,**The company can obtain customers' account history information through its online system and continuously update it, so that it can continuously delete new infringement threats. If potential infringement threats are detected, customers will be warned in a timely manner.

“When brand protection is automated, it takes an average of 10,000 hours from detecting an infringement to removing the infringing content.Four and a half hours, a process that would normally take days or weeks to complete manually,” Urquizu said.**Speed and accuracy are of the utmost importance.**And this can only be achieved through smart technology.”

Every day, Red Points' proprietary software is removed from the InternetMore than 100,000 casesIts technology currently helps more than 400 customers deal with fake content in more than 100 markets around the world, with a success rate of96%.

Rising awareness of domestic copyright protection

There are also such companies in China that are using artificial intelligence weapons to fight against piracy.

Since 2010, copyright issues have received increasing attention in China. In 2011, Wu Guanyong founded Guanyong Technology.**Cloud computing, big data and artificial intelligence technologies,**Create a big data platform for copyright monitoring, registration and protection.

There is also a company called Daodou Network located in Hangzhou.**All categories****The whole network monitoring comparison technology and artificial intelligence algorithm based on machine learning,**It provides content copyright and data services including copyright registration and confirmation, monitoring and evidence collection, rights protection, authorization management and marketing services to more than 50,000 individual creators and more than 150 companies.

In recent years, the copyright awareness of Chinese people has been gradually strengthened, but piracy and infringement are still rampant, and the rights holders have suffered greatly.

We hope that these AI anti-piracy technologies will be used by more individuals and businesses in the future. At the same time, the country will continue to improve relevant laws and regulations to make piracy and infringement incidents nowhere to hide.