The Los Angeles Times will show an AI-generated rating on political opinion pieces, the publication’s owner Patrick Soon-Shiong announced yesterday.
In a public letter, the billionaire owner said the new ‘AI Insights’ feature will not be applied to news stories but will be seen on some opinion pieces, columns, commentary and editorials.
The AI tool analyses the article’s written content and identifies where it falls on the political scale, offering readers an annotated summary of the ideas outlined in the piece and different viewpoints on the same subject.
The annotated summaries, named ‘Perspectives’, are generated in partnership with Gen AI chatbot Perplexity.
The political analysis, which the publication has called ‘Viewpoint’, tells readers where the article falls on the spectrum: Left, Centre-Left, Centre, Centre-Right or Right.
In the letter, Soon-Shiong outlines the AI tool uses a variety of factors to indicate the political essence of the article and considers: policy positions outlined, arguments used, whether there is a recognition of alternative views, the use of rhetorical devices, expert sources and government intervention, and offering an approach to solving the problem.
The publication said the analysis isn’t reviewed by anyone in the newsroom and aims to “only give readers additional perspectives”.
On X, Soon-Shiong celebrated the feature, calling the offering of alternative viewpoints a solution to an “echo chamber” – an environment where a person only encounters beliefs that align with their own.
Soon-Shiong bought the Los Angeles Times in 2018. Last year, he caused an internal stir after prohibiting the paper’s editorial team from posting in support of Kamala Harris during the US presidential election – which instigated a wave of resignations.
The paper’s journalists have expressed concern over the new AI tools recently announced.
In a statement, Matt Hamilton, vice chair of the LA Guild Times, said that despite backing efforts to improve media literacy and distinguish news from opinion articles, “We don’t think this approach – AI-generated analysis unvetted by editorial staff – will do much to enhance trust in the media. Quite the contrary, this tool risks further eroding confidence in the news. And the money for this endeavor could have been directed elsewhere: supporting our journalists on the ground who have had no cost-of-living increase since 2021.”
