Latvian journalists and editors: identifying AI-generated content has become a new responsibility

A survey conducted by Olsen+Partners shows that only one in five Latvian residents view the use of artificial intelligence in organizational communication positively. At the same time, 80% of respondents say they have noticed AI-generated advertisements, news articles, and social media posts. 

Latvian media professionals are also encountering press releases and opinion articles showing signs of AI use with increasing frequency in their daily work. Some journalists reject such materials outright, while others edit the text if the story itself is newsworthy. However, they are then more likely to view that organization's future communications with greater skepticism. What lessons from journalists' experiences should every communications professional keep in mind? 

1. Review AI-generated content carefully

"Last year, I almost made a serious mistake. I needed to create a chart based on material in which I had compiled a large dataset - around 400 different figures. I asked AI to process the data and generate a chart layout. It looked good and was easy to read, and I almost published it. Fortunately, I happened to notice one small error, which prompted me to carry out a thorough review. It turned out that 150 of the 400 figures contained mistakes. It was a terrible feeling! Since then, I no longer trust AI with tasks like this," recalls one of the journalists interviewed.

Other media professionals interviewed also emphasize that artificial intelligence can be a valuable tool, but blindly relying on AI-generated content is risky. A single clearly AI-generated email, social media post, or press release is enough to reduce trust in both the author and the organization. At the same time, news portal editors acknowledge that a new responsibility has emerged in their work: identifying AI-generated content, critically evaluating it, and deciding whether it is suitable for publication. 

2. Protect your voice and authenticity

"I've noticed that communication is becoming increasingly complex, with messages overloaded by bureaucratic or professional jargon. It often feels as though the people providing the information either do not want to be understood or fail to realize how difficult it is for the average reader to absorb and understand it without additional explanations, context, or a 'translation' into simpler language. I emphasize simpler,  not simplistic," explains one of the news editors interviewed.

Journalists acknowledge that AI is increasingly producing a uniform writing style grammatically correct, yet impersonal. When organizations publish such texts without editing them, they gradually lose their unique voice and what sets them apart from others. Over time, this can reduce both audience engagement and trust.

3. Let AI handle routine tasks, not decisions

Latvian media professionals most commonly use AI tools for translation, audio transcription, simplifying complex texts, and formatting content.

"I have created several AI agents, but I've designed them to handle tasks that AI can perform efficiently and without compromising quality, so I don't have to spend time on them. For example, when communicating with colleagues in Lithuania and Estonia, I use an AI-powered transcription and translation tool that converts audio into written text. I've configured it with specific requirements for the final output, including the text structure, sequence, job titles, and other important details," explains one of the journalists. 

At the same time, the interviewees draw a clear line: AI is an effective tool for handling technical tasks, but it cannot replace human judgment, editorial decision-making, or responsibility for the final outcome. 

"If a text is generated by AI and refined by a human or the other way around I don't see a problem with that. In my opinion, the biggest pitfall is relying on this technology 100%. It is not yet fully mature and, perhaps, never will be," says one of the journalists interviewed. 

"No one considers a calculator to be a mathematician. Likewise, artificial intelligence should not be regarded as a communications professional. Our research shows that the media are increasingly concerned not with whether an organization uses AI, but with whether its voice, positioning, and authenticity remain intact after AI has been used. If the hallmark of quality used to be the ability to use artificial intelligence, it will soon be the ability to use it in such a way that the text still feels as though it was written by a human," says Olga Kazaka, Partner and managing director of the communications and public affairs agency Olsen+Partners. 

In April and May 2026, the market research company Norstat conducted a public opinion survey on behalf of Olsen+Partners, interviewing 1,004 Latvian residents aged 18 to 74. This was followed by a series of 10 in-depth interviews conducted by Olsen+Partners with journalists and editors from Latvia's leading media outlets.