While still an unexplored field for the media in Moldova, artificial intelligence (AI) is gradually becoming part of the working tools of some journalists here. Media Azi reporters spoke to representatives of several media outlets to find out if and how they apply generative AI tools in their work or how reluctant they are.
The Northern Observatory team is using the Synthesia.io software as an experiment for the time being. Editor-in-chief Elena Cobăsneanu says she uses Synthesia for three tasks: creating templates for various columns, presenting news stories, and for video reports, which are the most popular. She also warns that the software can misread place names, for example, or put the wrong accent in speech. “It’s still a robot,” she says.
Some of the Agora portal team also use AI tools in their work. They mostly use ChatGPT for summarizing information and generating headlines, hashtags, and keywords used for social media. According to the media institution’s executive director, Irina Ghelbur, journalists were initially enthusiastic about it, but it is now met with reluctance from many journalists. Some fear they will lose their jobs and, as a precaution, try not to use it at all, says Alex Gurdila, general producer at Agora.
Alexandru Eftode, director of Radio Free Europe, points out that while speed is an advantage, it must be weighed carefully against the need to ensure that the information published or broadcast is accurate, objective, balanced, and meets the standards of quality journalism. According to him, colleagues on the team may use AI platforms for documentation or translation, but every response generated is verified.
Representatives of some media outlets are even more reticent. Ruslan Grabari, head of TV8’s news department, says the newsroom does not currently see ChatGPT, for example, as a reliable tool or partner. “We are afraid of errors. Besides, the information it provides cannot be verified, and the rules of journalism oblige us to quote live sources and not robots,” he says.
Victor Mosneag, editor-in-chief of Ziarul de Garda, points out that their newsroom does not use artificial intelligence at the moment either. “We are still studying the possibilities of using them more widely in our journalistic work,” he says.
Journalists from the Diez portal share the same opinion, saying they held a meeting with the team in April devoted exclusively to the topic of using artificial intelligence. “For now, we have decided not to use ChatGPT in the production of journalistic material. It’s not very stable yet and we can’t be sure of the information it provides,” says Petru Besleaga, editor-in-chief of the portal.
A study by the World Association of News Publishers suggests that half of news outlets use generative AI tools, while only 20% have defined ethical limits on their use.