The news communication industry is highly dependent on technology, and every new technology will take its transformation. Although the algorithmic news has just started, it has set off a revolution in the global media industry. KistianHammond, the Co-founder of “Narrative Science”which is a leading company in the field of algorithmic journalism in the Unites States, has predicted thatmore than 90 percent of the news will be generated by algorithmic robots in 2025. He further mentioned that, robots will soon win the honor of receiving the Pulitzer Prize which is traditionally awarded for human journalists.
There is no definition for algorithmic news. Some scientists name it Robot Journalism or Automated news and some even call it Data-driven journalism or Computational Journalism. However, in this article will refer to it as “Algorithmic News” since it highlights the core position of the algorithmic program in providing guidance for the news to be produced by the robot efficiently.
For traditional news, manin control of the the process of creating and publishing news. Journalists and editors rely on professional journalists’ perceptions, experience,experimental judgments and value preferences to select specific information and publish the appropriate news. For the algorithmic news,robot becomes the controller of the process. The news is generated by means of algorithmic tools and Big Data or large volume of Data which can generate new content automaticallyand helps to make the process more efficient.
Features of algorithmic news
When compared with the traditional ways and means of producing news, algorithmic news are comprised with three basic characteristics.
Firstly it is the leading edge of algorithmic technology. Whether it is robot news or data-driven news,the core technology is still the algorithm. It is the continuous improvement of the algorithm technology that makes the robot writing ability more and more close to human conventional level, producing news with high readability and gaining recognition from the audience.
Secondly it is the basic resource of the data. Algorithmic news is highly dependent on the data resources, especially on the quality, accuracy, effectiveness and real-time data. Therefore, algorithmic news can take the lead in the field of weather forecasting, stock market, sports events and so on.
Automated operations can be achieved in both release and promotion. A cost-effective and zero-cost breakthrough has been achieved in terms of cost. In summary, in the operating system of algorithmic news, the algorithm program is the core, the data resource is the foundation, and the intelligent operation is the key. Algorithmic news has been widely used in news production, but its danger is also obvious, which can effect the attitude of the audience and also changes traditional news values.
Thirdly it is the autonomous operation of artificial intelligence. In addition to the programs designed by humans, the algorithmic news can be automatically generated without any human intervention.The collecting information, analyzing data or even writing the news articles has now been replaced by Artificial Intelligence.
Risk of algorithmic news
1. Copyright disputes
Copyright disputes are the most pressing problem in relation to algorithmic news.
In China, there is a mobile application named ‘Today’s Headline’. As an information platform that starts with news aggregation and uses algorithmic technology to achieve personalized delivery, “Today’s headline” accurately grasps user’s needs and readingpreferences by using algorithmic technology. Within a few years’ time, this application became very popular in China. However, this application does not generate news but ratherengages in the distribution of news content, just like a ‘porter’ of news.
Unfortunately, this application had to face legal consequences for the same. Many news agencies sued the applicationin Court. Until August 2017, the number of cases in relation this application,accepted by Haidian Court alone exceeded 5,000 and it lost most of these cases.
2. Copyright of algorithmic news
As a news software for automatic writing, the news robot will acquire,ananlyze,refine data and ideas according to algorithm and automatically generate newsto deliver important informationto users speedily. The emergence and rise of robotic writing will have an immeasurableimpact on the production, editing, publishing or dissemination, distribution and promotion of news. At present, robot writing is used by world media. In China, Xinhua news agency have a robot journalist named “xiaoxin”, Tencent have robot journalist named “Dream writer” and Today’s headline have a robotic journalist named “Zhang Xiaoming”.
Among the many question that will arise in relation to intellectual property in this connection, most common are the following. Whoshall be the owner or the creator of the algorithmic news? Does the algorithmic news have copyright? Should the copyright belong to the robot or the institution that created the robot?According to the current intellectual property law, robot will have the copyright. But, robots are not entitled to rights of humans and they cannot exercise the same.Therefore, in others words, the copyright should belong to the creator or owner of the robot. On the other hand, if the robot make any mistake as to the news generation, who will take the responsibility for the robot? How can it be corrected?. If the basic data or algorithmic news is wrong, it is very likely that false news will be automatically generated and broadcastedspeedily.
3. ’Information Cocoons’ harm the right of information
The famous scholar Manuel Castells said that,power is based on the control of communication and information whether it is for the state or the media companies. The control of communication and information has always been the goal of various power competitions. It can be said that in the process of information exchange, power is controlled by the freedom of information. Especially, the field of news communicationbased upon human desire for freedom of information. The right to freedom of information is a universal right and is considered a fundamental human right in many countries.
The algorithmic technology is based on the large amounts of data.The emergence of algorithmic technology fully meets the demand in relation to point-to-point precision consumption in news products.It also allows the predictions of predecessors to be easily realized.This seems to be an improvement but it has also buried the hidden dangers, just as McLuhan said "in the beginning, we shaped the tools, and finally the tools in turn shaped us". Thus, the algorithm technology block the free flow of information.
Further, algorithmic technology has brought a reversed frog effect in the era of artificial intelligence which controls the exchange of information and results in ‘Information Cocoon effect’. The ‘cocoon effect’ means that people’s information reading habits will be guided by their own interests. If they continue to accept this information, they will enclose themselves in a standardized, procedural cocoon.
The user only sees what he wants to see, only hears the opinions he agrees with and the sounds with similar opinions are repeated which will make a relatively closed environment. These risks deprive individuals of their right to self-determination, undermine the diversity of information, manipulate the basis of information dissemination and in turn generate a monopoly of knowledge caused by mechanization brought about by the advancement of science and technology that Harold Innis is concerned about.
Although algorithmic technology controls communication rights it is only a gradual threat,but the free flow of information is the key to safeguarding human rights.
4. Infringement of privacy
Personalized and customized processes uses big data and algorithmic technology to discover and collect personal privacy. Therefore, with the development of algorithmic technology how to regulate privacy protection should be considered. For example, Facebook ‘likes’ can tell a lot about person. According to a study carried out by the American social media company in 2013,researchers showed they could use Facebook ‘like’ to correctly predict personal information, including a user’s gender and ethnicity.
The researchers could also predict a person’s age, intelligence, and opinions about religion and politics.
The study was based on information from 58,000 volunteers who provided “likes” to Facebook. The findings were published in the Proceedings of the National Academy of Sciences.Many companies urge people to log on to their websites with Facebook or other social media accounts. Thisin turn, provides companies with a complete picture of the user’s birthday, list of friends, schools attended and other personal information.Marketers often use Facebook ‘like’ and other digital records to sell products and improve services.
The researchers warned about the possible misuse of digital records and personal information. They said companies, governmental agencies, or even one’s Facebook friends could gather data that an individual may not have wanted to share.Therefore, Facebook should take necessary action to adequately protect personal information of their users and their friends.
Regulation of algorithmic news
Based on the principle of "Center of the audience", the algorithmic technology satisfies the user's subjective needs by directly collecting the user's preferences. This personalized and customized news and information often ignores the importance of information and the significance of information. In other words, when the algorithmic technology uses the accumulated user's interest as a push guide, the media platform also mistakes the user's ‘attention’ of the specific information as ‘importance.’ Obviously, this is unfair to users and it is also completely contrary to the spirit of the Internet.
How to regulate algorithmic news?
Firstly, the social norms of algorithmic technology in news depend on the self-discipline of the enterprises. For example,there is a need to invite academic experts in the fields of law, communication, society, and ethics to form a research committee to increase the analysis and judgment of algorithmic technology cases, and actively accept the management and guidance of relevant government departments to improve the rules of algorithmic technology.
Secondly, there is also a need to rely onthe supervision of the user. For example, the mobile application “Today's headlines” have been launched on the software named “Ling gou” and this software could detect and filter information to be anti-vulgar, anti-pornography, anti-false information. When the user is using the relevant text or link, the ‘Ling gou’ will judge the health of the related content. According to reports, more than 600,000 users in China have used“Ling gou” software to filter the internet platform.
Thirdly and importantly there is a necessity to build an algorithmic association. In order to improve the application standards and devise and impriveoperating rules an algorithmic association should be formed. The rules formulated by such an association in terms of algorithmic technology will improve the industry’s self-discipline ability.
Finally, a third party “Ethics Committee” could be established to find problems and deficiencies and to address them in order to promote algorithmic self-discipline and ethics of algorithmic technology.
(Dr. Zhao Ying is Associate Director of China-Sri Lanka Cooperation Studies Centre and visiting scholar of Southwest University in Chií and Dr. Cai Fei, Associate Professor of Southwest University of Political Science and Law in Chií. Comments are welcome at: firstname.lastname@example.org/www.pathfinderfoundation.org)