How AI is Changing Journalism
What’s this about: The view that journalism will be heavily impacted by AI is held all around the globe. In 2019, the London School of Economics published a global survey of journalism and AI based on interviews with 71 newsrooms in over 32 countries. One of the conclusions was that AI will reshape journalism, at first in an incremental way, but there will be long-term structural effects. The technology is set to disrupt the field in many ways involving data analytics, prediction making, plagiarism detection, NLP, fake news detection, and more.
Go Deeper to Learn More →
Back in December 2019, The Economist published an interview with the natural language processing (NLP) model GPT-2, which demonstrated how the model could answer questions about the general state of the world.
Question posed to GPT-2 model: “What is the future of AI?”
Response: “It would be good if we used the technology more responsibly. In other words, we should treat it like a utility, like a tool. We should put as much effort into developing the technology as necessary, rather than worrying that it’s going to harm us and destroy our lives.”
While GPT-2 at that time did not ‘understand’ what it was saying, the implications were huge nonetheless. It demonstrated that in the near future, AI bots could write and generate text that is indistinguishable from a human being.
AI Uses in the Field
AI is deployed on many different fronts in journalism, whether it be to create news stories through NLP, detect plagiarism, perform data analysis, or completely take over as news anchors.
Let’s take a look at some of the specific ways AI technology has been used in the field:
Data Analysis and Prediction Making: No matter the field, one of the most valuable aspects of AI is its ability to analyze large datasets. Bloomberg News uses AI robots to analyze this data and alert journalists of trends or anomalies. Humans are estimated to create 1 quintillion bytes of data every day, meaning it is impossible for journalists to sift through this amount of information.
Plagiarism Detection: AI can help detect plagiarism in articles by mining data, analysis, and transcribing interviews. Plagiarism is one of the main concerns of every journalism and news agency, since it can ruin careers and the reputation of the agency at large.
Fake News Detection: In a world full of fake news, AI systems can help differentiate between real and false content, helping news agencies avoid potential scenarios that are damaging to everyone.
AI Tools in Journalism
Perhaps the most exciting aspect of AI in gaming for everyday players is its ability to enable developers to Not many people know this, but many of the major news organizations and the field of academia have already developed specific AI tools. As time goes on, these tools will be increasingly implemented to both supplement human journalism and completely take over some tasks. This means journalisms are going to adapt to the technology, working alongside it just like other industries such as manufacturing.
New York Times R&D Lab: The NYT has a research and development lab focused on exploring emerging technologies involving computer vision, photogrammetry, spatial computing, media transmission, NLP, and misinformation. Misinformation, which we detailed in ‘The Rise of AI-Enabled Disinformation,’ is becoming one of society’s greatest threats.
Washington Post: The Washington Post has experimented with news discovery tools, including one called Lead Locator, which is specifically aimed at supplementing national politics by suggesting interesting locations for reporters to report on.
Academia: In the world of academia, conferences like Machines +Media and the Computation + Journalism symposium have been established to explore AI technology in journalism. The development of natural language processing (NLP) models are also key, with AI researchers at the Beijing Academy of Artificial Intelligence recently announcing Wu Dao 2.0, the world’s largest NLP model that is ten times bigger than GPT-3.
Will AI Replace Our News Anchors?
Back in May 2020, the Chinese state news agency Xinhua released its latest version of an AI 3D news anchor, which joined an expanding list of virtual presenters being developed by the agency. Named Xinn Xiawei, the AI news anchor was based on one of the agency’s real-life human anchors, and according to co-developer Sogou, it utilizes “multi-modal recognition and synthesis, facial recognition and animation and transfer learning.”
Take a look at the oddly dystopian vibe given off as the agency’s first AI news anchor gives an update:
The AI anchor developed in China is extremely complex but still in its infancy, meaning we could be headed toward a world where AI news anchors are virtually indistinguishable from humans.
This poses the following questions: Is this where we are headed? And if so, would you feel more inclined or less to gather news from an AI anchor? Would you rather your news anchors be totally void of emotion and strictly operating on facts, or would you rather them still be in touch with human emotions?
Savior or Destroyer of Journalism?
As of right now, artificial intelligence does not look to be on a path of complete disruption for the field of journalism, unlike other fields. With that said, while we should not be worried about a robot takeover of journalism, news organizations must take the correct steps when implementing the technology.
A 2017 study carried out by the Tow Centre of Digital Journalism suggests that AI technologies should have editorial values integrated into their design. This would provide transparency to readers who are taking in AI-enabled news, as this type of design would tell us exactly how a story was put together with the technology.
Over in the EU, the Council for Mass Media in Finland published a report recommending that the media council should immediately take up the issues of data processing, choices in computerized procedures, and transparency. If this does not happen, the report warns that others will take over and possibly “jeopardize the freedom of press.”
If the media does not act in an ethical and transparent way, it could have a devastating impact on the public’s faith in journalism. And with the public’s faith in such institutions at an all time low, it is crucial for the media to proceed very cautiously. So to answer that question of savior or destroyer, only time will tell. It will largely depend on how the technology is implemented, which is the case for virtually every industry set to be disrupted.