Nieman Lab.
Predictions for
Journalism, 2024.
In 2024, the ascent of AI in news won’t lead to a Cambrian explosion of articles written by souped-up autofill algorithms. Instead, we’ll see the beginnings of ultra-segmented news presentations.
This evolution will be driven by the genre-redefining changes in search engines we’re seeing today: AI summaries of query results that make clicking through to sources increasingly unnecessary.
We’ll see publishers turn away from SEO and toward building direct relationships with readers. Journalists will continue writing inverted-pyramid style news reports. (The summarization AI will have been trained on these, after all.) However, for an increasing subset of readers, “articles” will be as invisible as CSS code.
The articles journalists write will become source material to be remixed and delivered in hyper-personalized presentations. (One easy example is that most people likely prefer to read the news in their native language). Building extremely personalized news products doesn’t require technology beyond what today’s LLMs already offer: It’s only lacking innovative leadership and funding to start growing the idea.
We’ll see the rise of a new journalistic craft that could breathe life into the industry: These first experiments will probably come in the form of email newsletters constructed from pieces of articles, graphics, and linked source material — remixed to fit a particular reader’s interests and level of context with the help of AI.
With the rise of personalization, we’ll see the beginnings of a new journalism craft: news DJs. Their main job will be to edit these automated news presentations. But they will also need to artfully manage the give and take between business prerogatives, machine learning engineers’ priorities, and the ethics of news presentation. This wide range of responsibilities will likely inspire better-heeled newsrooms to hire editorial AI strategists to lend a hand.
Over time, some prominent reporters will retain their direct connection to an audience — but this will be increasingly rare. Instead, reporters will be deployed as elite creators of trusted source material. The economics here might resemble that of tech programmers today: well-compensated and largely unknown to the customer.
The news DJs, on the other hand, will be tasked with managing direct relationships with small subsets of a publication’s audience. They’ll probably even pass requests for information on to reporters and communicate directly with readers if they can get an answer. More likely, though, the reporter will point to the roadmap and ask them to file a ticket.
Bassey Etim is a journalist and cofounder of the AI-in-media newsletter Machines on Paper.
In 2024, the ascent of AI in news won’t lead to a Cambrian explosion of articles written by souped-up autofill algorithms. Instead, we’ll see the beginnings of ultra-segmented news presentations.
This evolution will be driven by the genre-redefining changes in search engines we’re seeing today: AI summaries of query results that make clicking through to sources increasingly unnecessary.
We’ll see publishers turn away from SEO and toward building direct relationships with readers. Journalists will continue writing inverted-pyramid style news reports. (The summarization AI will have been trained on these, after all.) However, for an increasing subset of readers, “articles” will be as invisible as CSS code.
The articles journalists write will become source material to be remixed and delivered in hyper-personalized presentations. (One easy example is that most people likely prefer to read the news in their native language). Building extremely personalized news products doesn’t require technology beyond what today’s LLMs already offer: It’s only lacking innovative leadership and funding to start growing the idea.
We’ll see the rise of a new journalistic craft that could breathe life into the industry: These first experiments will probably come in the form of email newsletters constructed from pieces of articles, graphics, and linked source material — remixed to fit a particular reader’s interests and level of context with the help of AI.
With the rise of personalization, we’ll see the beginnings of a new journalism craft: news DJs. Their main job will be to edit these automated news presentations. But they will also need to artfully manage the give and take between business prerogatives, machine learning engineers’ priorities, and the ethics of news presentation. This wide range of responsibilities will likely inspire better-heeled newsrooms to hire editorial AI strategists to lend a hand.
Over time, some prominent reporters will retain their direct connection to an audience — but this will be increasingly rare. Instead, reporters will be deployed as elite creators of trusted source material. The economics here might resemble that of tech programmers today: well-compensated and largely unknown to the customer.
The news DJs, on the other hand, will be tasked with managing direct relationships with small subsets of a publication’s audience. They’ll probably even pass requests for information on to reporters and communicate directly with readers if they can get an answer. More likely, though, the reporter will point to the roadmap and ask them to file a ticket.
Bassey Etim is a journalist and cofounder of the AI-in-media newsletter Machines on Paper.
