The Shiny New Toy No One Knows How to Use: Why Is AI in the Newsroom So Tricky?
Hi everyone, John here! For a while now, we’ve been hearing that Artificial Intelligence, or AI, is going to completely change the world, including how we get our news. The idea was that news companies would use this powerful new technology to do all sorts of amazing things, almost overnight. But if you look closely, that big, flashy revolution hasn’t quite happened yet. It’s a bit like everyone got a super-advanced, futuristic kitchen gadget for Christmas, but the instruction manual is missing.
So, what’s really going on? Why are news companies, who are so excited about the potential of AI, finding it so difficult to actually put it to good use? Today, let’s break down this puzzle. It’s a fascinating story about hype, reality, and the hard work of real change.
Welcome to the “Trough of Disillusionment”
When a big new technology comes along, it often follows a predictable pattern. First, there’s a huge wave of excitement. Everyone talks about how it will solve every problem. This is the “peak of hype.” Then, reality sinks in. People realize it’s more complicated and less perfect than they imagined. They get frustrated, and the excitement cools off. This dip is called the “trough of disillusionment.”
Lila: “John, that’s a funny name. What exactly is a ‘trough of disillusionment’? It sounds a bit sad!”
John: Haha, it does, Lila! But it’s actually a normal and even healthy part of the process. Think of it like this:
- The Peak: You hear about a new video game that looks incredible. You think it’s going to be the best game ever made!
- The Trough: You finally play it, and the controls are tricky, there are some bugs, and it’s much harder than you expected. You feel a bit let down and might even stop playing for a while. That’s disillusionment.
- The Comeback: But then, you learn the controls, figure out the strategies, and start appreciating the game for what it is. You find the real fun in it.
That’s what’s happening with AI in the news business. After the initial wild excitement about Generative AI, companies are now in that tough middle part, figuring out the tricky controls and how to actually make it work for them.
Lila: “Okay, another term! What makes ‘Generative AI’ special?”
John: Great question! Most older AI was good at analyzing things that already exist. Generative AI is different because it can create, or ‘generate,’ new things. It can write an e-mail, create a picture, or summarize a long report, all from a simple instruction. It’s the technology behind tools like ChatGPT, and it’s what got everyone so excited because it feels like you’re talking to a creative assistant.
The Three Big Hurdles for News Companies
So why are news companies stuck in this “trough”? It turns out there are three main challenges they need to overcome. It’s not as simple as just buying some AI software.
Hurdle #1: The People Problem (Talent)
You might think the solution is just to hire a bunch of brilliant AI experts. But it’s not that easy. An AI coder might be a genius at technology, but they probably don’t understand the pressures of a news deadline, the importance of an anonymous source, or the ethics of journalism. On the other hand, a veteran journalist knows the news business inside and out but might not know the first thing about how an AI model works.
The challenge is finding people who can bridge this gap. News organizations need teams that have both technical skills and deep newsroom knowledge. It’s like building a championship-winning sports car. You don’t just need a great engineer and a great driver; you need them to be able to talk to each other constantly to get the best performance out of the car.
Hurdle #2: The Technology Isn’t a Magic Button
AI isn’t a simple app you can download on your phone and it just works. To be truly useful, it has to be connected to all the other systems a newsroom already uses.
Lila: “The original article mentioned something called a ‘CMS.’ What’s that, and why does it matter?”
John: Excellent question, Lila. A CMS stands for ‘Content Management System.’ It’s the main software tool journalists use to write, edit, and publish articles online. For our blog, we use a CMS called WordPress! For an AI tool to be helpful, it needs to connect smoothly with that CMS. For example, if you want an AI to suggest three different headlines for an article, it needs to be able to ‘see’ the article inside the CMS and then ‘paste’ its suggestions in the right spot. Getting all these different technologies to talk to each other is a huge technical puzzle.
There’s another, even bigger technology problem: AI can be wrong. Sometimes, an AI will confidently state something that is completely made up. In the tech world, they have a spooky name for this.
Lila: “Wait, I remember reading this… they call it ‘hallucinations’? Does that mean the AI is dreaming?”
John: Haha, not exactly, but it’s a good way to think about it! An AI ‘hallucination’ is when the AI generates false information but presents it as a fact. It does this because it’s designed to predict the next most likely word in a sentence, not to check if the sentence is true. For a business like news, which is built entirely on trust and accuracy, this is a massive problem. You can’t have your helpful AI assistant inventing quotes or making up statistics. This is a major reason why newsrooms are being so cautious.
Hurdle #3: The Biggest Challenge of All (Transformation)
This is the toughest hurdle. Technology is one thing, but changing how an entire company of people works is another. You can’t just create a small “AI innovation team” in a corner and expect the whole organization to change. The use of AI has to be woven into the daily jobs of reporters, editors, social media managers, and everyone else.
This requires a clear plan from the very top leaders of the company. They need to decide: What is our goal with AI? Are we trying to save time on boring tasks? Are we trying to create new types of stories? Are we trying to help readers find articles they’ll love?
Without a clear strategy, AI just remains a shiny toy. It’s the difference between giving one person a power drill and teaching an entire construction crew how to build a house using new, more efficient power tools. It requires new blueprints, new training, and a new way of thinking for everyone involved.
So, How Can Newsrooms Climb Out of the Trough?
The situation isn’t hopeless! The article points out a clear path forward. Companies that are starting to see success are doing a few key things:
- Spread the knowledge: Instead of keeping AI experts in a separate bubble, they are integrating them into different departments so everyone can learn.
- Start small and be specific: Don’t try to solve everything at once. Focus on one or two specific, nagging problems. For example, can AI help summarize long government reports to save a reporter hours of reading? Can it create quick social media posts from a finished article?
- Leadership must lead: The bosses need to champion the change and communicate a clear vision for how AI will help make everyone’s job better and serve the reader more effectively.
- Focus on assistance, not replacement: The most successful uses of AI are ones that act as a helpful co-pilot for journalists, taking care of tedious work so they can focus on what humans do best: investigating, talking to people, and telling compelling stories.
A Few Final Thoughts
John’s View: To me, this all makes perfect sense. News is a business built on trust, so it’s right to be cautious with a technology known for making things up. The slow, careful approach isn’t a sign of failure; it’s a sign of responsibility. This is a marathon, not a sprint, and newsrooms are wisely taking their time to get it right.
Lila’s View: As someone new to all this, it’s actually a relief to hear that it’s not some magic that I’m supposed to understand instantly. Knowing that even the big companies are still figuring it out makes it all feel less intimidating. I think it’s really exciting that AI could help journalists with the boring stuff, so they have more time for the really important stories that affect us all!
This article is based on the following original source, summarized from the author’s perspective:
Are news companies in the era of unfulfilled potential with
AI?