Journalists’ main skill was writing. Not any more.
AI can write, and increasingly it can write well. Agents can learn your style, transcribe interviews, interrogate hundreds of pages, find inconsistencies, research deeply, scan your competitors and test hypotheses.
That changes the value of the journalist.
The journalists’ skill that matters now is judgement. Perhaps it always was. What is the angle? What doesn’t add up? Is the source credible? What is missing? Which fact matters? What question comes next?
A good journalist develops a feel for these things over years: thinking and experience, curiosity and scepticism. At its best, wisdom.
And does the audience really care what role AI plays? As long as the journalist signs off.
On deadline: Writing is no longer a key journalistic skill (Midjourney)
AI makes that distinction clearer
Writing is the visible product. Behind a good story is something more important: deciding what is worth writing.
AI emphasises the distinction.
It can read the 200-page report. The journalist decides which paragraph matters.
It can suggest six questions for the chief executive. The journalist spots the hesitation after question six and asks the seventh.
It can summarise both sides. The journalist decides whether the evidence gives them equal weight.
It can find an anomaly in thousands of financial records. The journalist decides whether it is noise or a story.
Caution is justified. AI invents, misunderstands and produces plausible nonsense. Journalists must verify. Publishers must protect credibility.
But what if being too careful is dangerous too?
There is a difference between protecting what you publish and restricting what journalists can learn.
At publication, ensure credibility: check every fact and quote; make the journalist and editor accountable.
Before publication? Experiment. Why wouldn’t you? Find out where AI is brilliant. Find out where it is stupid. You learn AI by using it.
The Author – Peter Wilkinson
And that creates a new divide.
I talk to journalists as part of my work almost every day. Curious, I ask them about AI. Some are trailblazers; many trail.
Journalists who experiment every day learn faster than those who don’t. Intensive users learn to frame problems, supply context, interrogate answers, spot hallucinations and redesign reporting around new capabilities.
The gap widens. Those who hold back fall further behind, as does their newsroom. And their audience.
And trailblazers must keep learning. AI does not stand still. The journalist who mastered it two months ago but stopped experimenting is already falling behind.
When advanced users in other sectors are beginning to ask, “When should our agents be required to stop and ask a human?”, publishers and editors should, at least, be asking, “Are our people learning to manage AI agents faster than our competitors?”
The paradox: AI makes judgement more valuable
The more AI can do, the more valuable the human contribution becomes.
Reuters Institute has reported that technology companies are employing journalists to train AI systems. What are they teaching the machines? Accuracy, clarity, and natural structures.
AI makes routine production cheaper and judgement more valuable.
A first draft matters less. Knowing whether it is true and the final edit matters more. Curiosity, scepticism, courage, knowledge and judgement rise in value.
AI does not diminish the experienced journalist. Used well, it amplifies one.
A different kind of journalist
Working with AI is collaborating with a robot. Weird? Perhaps, especially for journalists.
Journalism celebrates the maverick: working alone, chasing the exclusive, competing for the front page. Yet some of my best journalism was collaborative: producer and reporter; reporter and researcher; reporter, editor and executive producer.
AI is just another collaborator. Seen that way, AI is not a radical break with journalism. It is a new assistant. Not a big deal.
That suggests two rules for news organisations:
- Be conservative about what you publish with AI.
- Be adventurous about how your journalists use it.
Protect standards at publication. Before that, let journalists experiment, build agents, redesign workflows and fail safely.
Otherwise, the gap between the leading edge and the laggards widens.
Journalists lose as their skills fall behind. Publishers lose as their newsrooms become less capable. Readers lose when journalists spend time doing work machines can do instead of the work only good journalists can do.
Become an AI native now
We talk about digital natives. AI natives are next: people for whom working with artificial intelligence is simply how work gets done.
But why wait for the next generation? Become an AI native now.
As intelligence is always available, experiment constantly. Question every workflow. Ask not, “How can AI help me do this?” but, “Given what AI can do today, how should I do this now?”
Judgement without adaptation isn’t enough. The experienced journalist who combines judgement with an AI-native approach has the advantage.
