PETALING JAYA (May 14): Just as the atomic bomb ushered in the nuclear age, generative artificial intelligence’s (gen AI) potential singularity looms over journalism. With growing mistrust towards journalism and AI posing as a potential threat to the industry, crafting ethical guidelines for Malaysian newsrooms becomes paramount.

Gen AI is found to be overwhelmingly used for content creation in newsrooms, with 69.6% of organisations using gen AI to craft news headlines, social media posts, newsletters, taglines, and story drafts according to the generative AI in journalism report done by the Associated Press. 

“AI crawlers can go through news and trends online, and produce an article to be sent to the editor. Then, ChatGPT will do the editing and generate the graphics, while the publishing of the article can be automated as well. This replaces a large part of the process for journalists,” said Azeem Abu Bakar, the managing director of Free Malaysia Today (FMT) News, during the Journalism in the Age of Artificial Intelligence roundtable held at the University of Nottingham Malaysia Learning Centre on May 9.

AI crawlers, also known as web spiders, are software programmes designed to automatically browse websites and extract data. This data is then used to train and improve AI models developed by various companies.

With local media companies suffering a decline in revenue, Azeem emphasised the need to evolve and transform their business models to avoid being shut down, as market consolidation is inevitable.

“But there will always be a need for news by editorial teams, because the integrity of information is paramount to ensure that we get the right information for the sake of our democracy,” said Azeem. 

As such, investigative journalism and reporting political intricacies demand the utmost integrity in information. While gen AI excels at data analysis, journalists' strength lies in uncovering hidden patterns and engaging relevant stakeholders, which are distinctly human skills crucial to the industry's survival.

“[The media companies] that will survive will be those with compelling storytelling, explainers of stories that can shape public opinion, and rich lifestyle content that can build brands. You’ve got to be sophisticated [in your writing],” said Azeem. 

Azeem also shared that media for equity deals could be a strategic way for media companies to invest in promising start-ups or brands. 

By offering advertising space in exchange for equity, media companies foster long-term relationships with these emerging businesses, potentially benefiting from their future growth through increased revenue from both advertising and potential equity buyouts.

“Media companies, with their influence, can increase the valuation of another company. I don't think AI is able to do that, [compared to] journalists who are able to connect the dots in the past and bring in relevant people to advocate for policy, which can potentially save an entire industry,” said Azeem. 

Of course, that is not to say that media companies should completely forgo the idea of using gen AI in newsrooms. Transparency about its use in writing, particularly with established guidelines, is essential to building reader trust.

“Transparency plays a huge part in being able to trace accountability. The Associated Press describes in detail the AI process, which helps encourage transparency. In all stories, we include disclosures that state this story was generated using AI,” said Ernest Kung, the AI product manager for the Associated Press. 

Conversations surrounding the usage of gen AI in journalism needs to be had, especially on crafting guidelines, because it is not going to be a one-size-fits-all solution, said Shafizan Johari, the chief of multiplatform news innovation at Astro Awani. 

Communicating where AI processes are taking place, coupled with transparent guidelines, will aid newsrooms in keeping up with the changing landscape of journalism. However, there is still one elephant in the room that needs to be addressed, that is the control of the state over media companies.

“The growing sense of distrust and discontent is not just aimed at politicians, but also spreading horizontally within society, coupled with the rise of AI, which adds another layer of complexity to the current social climate,” said Ashwad Ismail, the editor-in-chief of Astro Awani.

The conversation around media freedom often focuses on control. However, the key issue might lie in the access to public data. With open access, media companies could leverage machine learning to analyse trends, and potentially predict future developments over the next decade, said Ashwad. 

“When you have the state imposing certain values or providing censorship onto creative media products, it puts people off. No matter how much you want to produce, if those restrictions are there, you are not going to go very far,” said Professor Zaharom Nain, a professor of media and communication studies at the faculty of arts in the University of Nottingham Malaysia. 

Malaysia is still lacking regulations and discussion surrounding the transparency of AI utilisation in journalism. Ultimately, it is on humans to ensure its ethical use. Media companies are encouraged to start collaborating with one another to facilitate the necessary discussions about regulating AI and ensuring ethical use of it.

In efforts to harness the capabilities of AI while ensuring the quality and credibility of news reporting is not compromised, Astro Awani in collaboration with the University of Nottingham Malaysia and the Centre for Independent Journalism Malaysia (CIJ) launched Jiwa, guidelines for AI utilisation in Astro Awani. 

The guidelines aim to address key areas of concern in the usage of AI in the journalistic process, while preserving the sanctity of journalism.