top of page
Search

AI and Journalism: What Gets Faster… and What Gets Lost

  • Writer: Joshua J Smith
    Joshua J Smith
  • Mar 31
  • 4 min read

AI tools like ChatGPT, Perplexity, Gemini, and Claude are rapidly becoming embedded in how news is produced, not just how it is consumed. In journalism, this raises a different kind of question:


What happens when efficiency becomes the default standard for storytelling?


A humanoid robot with glowing blue eyes surrounded by digital screens in a futuristic setting. Newspapers, glasses, and pens in foreground.

In my MASC 311: AI in Mass Media course, this tension shows up almost immediately when we reach the journalism unit. Some students are still unsure whether journalists should be using AI tools at all. The instinct is understandable. For many, it feels like a shortcut that conflicts with ideas of originality, credibility, and verification.


I often pause there and introduce a broader frame. I pull up the 2023 Tow Center report from Columbia Journalism Review to show that AI in journalism is not just about writing articles. It spans the entire production process. For example, AI is used for verification and fact-checking, transcription of interviews, and even searching archives to surface relevant context. And, bonus for translating your story into another language for a group that might otherwise struggle to access news.


Once students see that, the conversation shifts. This is no longer about replacing journalists. It is about augmenting the workflow.


Chart detailing AI applications in news, from access and observation to publishing. Tasks include information discovery and content moderation.
From Tow Center for Digital Journalism. (2023).

There is usually a moment of recognition, and sometimes a bit of discomfort. It reminds me of the old reference point many journalism programs used, the film Shattered Glass (2003), where the lesson centered on fabrication, verification, and editorial oversight. The fear then was that a reporter might invent details. The concern now is different. It is not that AI is fabricating in the same way, but that it can produce something that appears complete while quietly lacking depth or context.


This distinction becomes clearer when we look at the research.

A recent systematic review by Sonnie, et al (2024) of 127 studies on AI in journalism found that:


  • Speed increases in 76% of cases

  • Accuracy increases in 58% of cases

  • But that:

    • Context declines in 42% of cases

    • Depth of analysis declines in 32% of cases


This is where the implications begin to extend beyond the newsroom.

If AI is reshaping how news is produced, it is also reshaping how it is encountered. The same systems that assist with writing, editing, and distribution are increasingly influencing how audiences discover and engage with information. Not just behind the scenes. Right at the point of access.

That shift matters.


AI and Journalism Audiences are Changing (Too)

Research supports this change. A recent report from Muck Rack’s Generative Pulse analyzed more than one million links cited in AI-generated responses and found that 82% of those links come from earned media sources, with 94% coming from non-paid media overall. (See my other post about the future of PR and earned media). And, the ongoing lawsuit filed by The New York Times against OpenAI underscores growing tensions over how AI systems use journalistic content, raising critical questions about copyright, attribution, and the future value of original reporting in AI-generated outputs.


News consumption is shifting from direct engagement with outlets to mediated experiences through AI systems, search summaries, and curated feeds. Audiences increasingly encounter news as condensed, personalized outputs rather than full articles, often without visiting the original source. As a result, journalism is still informing the public, but the pathway between publisher and audience is becoming less visible and more algorithmically controlled.


Journalism can come out ahead, but only if it leans into what matters most.

Right now, just 28% of Americans say they trust the media to report the news fully and fairly, and while about 56% report at least some level of trust, that number has been declining over time. This is not just a credibility gap. It is an opportunity.

As AI accelerates the production and distribution of information, the differentiator is no longer speed. It is trust, depth, and clarity. News organizations that prioritize original reporting, transparent sourcing, and meaningful interpretation will not just compete in an AI-driven environment. They will define it.


So I suppose it is that old tale of the tortoise and the hare: In a system optimized for speed, journalism’s advantage may come from those willing to slow down just enough to add context, verify meaning, and earn trust.


Joshua J Smith, PhD

By Joshua Smith, PhD

Co-Founder, Chair, Media + AI



AI Disclosure:

Portions of this post were developed with the assistance of generative AI tools (Claude/Gemini) to support drafting and editing. The author reviewed, revised, and verified all content, analysis, and sources to ensure accuracy, originality, and alignment with the cited materials.


Image: Google Gemini

Comments

Commenting on this post isn't available anymore. Contact the site owner for more info.
bottom of page