We’re halfway through the year, so it feels like a good moment to do a bit of a pulse-check on AI use cases in the media industry. Where are news publishers using AI and what are the initial outcomes? 

It’s no surprise to anyone working in the industry that finding efficiencies is top of mind for many publishers — so, when they find efficiencies, how are they reallocating their resources? How are AI practitioners feeling right now and how are news consumers thinking about AI?

Let’s dig in!

Sonali

The era of doing more with less — and unfulfilled potential

Where is AI being used in news organisations?

Workflow automation tops the list, followed by video and image optimisation, according to a survey of 108 publishers and broadcasters undertaken by Digiday and Arc XP.

But here’s some food for thought: If you look at which tools aren’t working well for publishers, the list is led by video and image optimisation, content recommendation and distribution, and real-time data and alerts.

“As for the ways publishers and broadcasters are using AI so far, improving workflow efficiency tops the list (94%), followed by automating scheduling and distribution of content (82%) and automating video editing (i.e., social clips, adding subtitles) (82%),” the report said.

As many as three-quarters of news organisations are using their AI budgets on editorial and content creation, the report said. In fact, 93% of respondents said they had the editorial/content team handling AI initiatives, with an overlapping 90% said they had a dedicated AI-focused team that collaborated with many departments.

How about using AI for customer-facing products? The top use cases were automated creation of accessible content (88%), personalised content delivery (85%), and automated fact-checking (84%).

News publishers are also using AI to deepen engagement with their audiences via tools that monitor and moderate comments (88%) and predict emerging topics (85%).

“As AI automates tasks and works on menial items that require a lot of time, publishers are dedicating more efforts to strengthening editorial oversight (86%), spending more time connecting with local communities and audiences (85%), and focusing on innovation and experimentation (76%).”

But what is also significant is what they are generally not moving resources into: an increased focus on training and upskilling (19%), amplifying underrepresented voices (18%), or building revenue and growth strategies (12%).

This sentiment was echoed in another survey, which pointed out that the media industry is largely playing it safe and using AI for tactical functions rather than for bold, ambitious purposes.

“AI can save time and reduce friction, but it hasn’t yet earned trust as a creative or strategic tool,” according to the other survey, of 131 professionals undertaken by The Rebooting and BlueConic. “Teams are using it to reduce repetitive tasks, accelerate production timelines, and support experimentation without hiring. Still, most see it as a supplement, not a replacement.”

“There’s no AI strategy,” the report cites an executive as saying. “There’s just trying to do more with what we have.”

The focus is on utility, it points out, as the media industry settles into a permanent state of doing more with less.

“The teams that are gaining traction are those that have found alignment between what audiences value and what the business can support. They’re not building flashy features to impress stakeholders — they’re building useful products that deepen engagement, capture first-party data, and support sustainable revenue.

“If there’s a prevailing mood around AI inside media companies, it’s one of cautious optimism strained by operational disarray. Nearly every publisher we surveyed or spoke with acknowledged that AI has the potential to reshape their business, but most also admitted they haven’t figured out how to turn potential into execution.”

GenAI can help solve problem of news avoidance

How does all of this line up with news consumers’ expectations? 

The Reuters Institute for the Study of Journalism at Oxford University has released its annual mammoth survey of almost 100,000 people in 48 different markets across the world. 

Friends, I read through all 171 pages of their report so that you don’t have to and pulled out some key insights on where AI applications can be helpful.

1. AI chatbots are being used as a source of news for the first time, particularly among younger audiences, the report notes. “As the largest tech platforms integrate AI summaries and other news-related features, publishers worry that these could further reduce traffic flows to Web sites and apps. But we also show that in a world increasingly populated by synthetic content and misinformation, all generations still prize trusted brands with a track record for accuracy, even if they don’t use them as often as they once did.”

2. Video continues to grow in importance as a source of news. (You can use GenAI for text to video.) Across all markets, the proportion consuming social video has grown from 52% in 2020 to 65% in 2025. In the Philippines, Thailand, Kenya, and India, more people now say they prefer to watch the news rather than read it.

3. Podcasts are an increasingly important way to reach younger, better-educated audiences. (You can use GenAI for text to podcast.) To attract the next generation of subscribers, publishers have started to experiment with a range of payment options that include early access, extra content, and even separate subscriptions at a lower price point. About 42% of news podcast listeners across 20 countries say they would be willing to pay a reasonable price for news-related podcasts they like. A high proportion of people (73%) also say listening to podcasts helps them understand issues at a deeper level; 71% of news podcast listeners are extremely or very interested in news compared to 45% of non-podcast listeners. 

4. Personalisation of content is gaining traction. “There is some enthusiasm for making the news more accessible or relevant.” (You can use GenAI for personalisation.)

5. Addressing news avoidance. Take a look at this chart. Look at all the reasons people avoid the news:

Now, think of how AI can help tackle these problems:

• A solutions journalism bot can suggest positive ways to write a story. 

• A personalised, AI-driven feed or alerts strategy can cut down on the amount of news we push to people, particularly on certain topics they are exhausted by. 

• Data-driven insights powered by AI and using GenAI to change the format of the news can make it more relevant to the consumer’s needs. 

• GenAI can certainly rewrite the news to make it more comprehensible to people from different educational and linguistic backgrounds.

“We asked respondents about their interest in eight different potential AI applications that could be used to better suit their individual needs. We found the interest in AI personalisation to be highest when it comes to approaches that make news content quicker/easier to consume and more relevant, such as summarised versions of news articles (27%) and translations of news articles (24%),” the report said.

In other words: Many of us are still writing articles that are too long and that assume the reader knows the essential context of the news story. We can use GenAI to present the same information in different formats so our audience has a choice in how they consume and process the news.

“More broadly, we find that interest in options for adjusting the format and style of news is higher than options for personalising the selection of stories likely linked to those concerns about missing out.” 

This is interesting because, let’s face it, the news industry has spent at least 15 years working on recommendation engines that select which articles should be placed in front of readers.

What else are we getting wrong? 

“The danger is that publishers will use automation to cut costs and chase new AI algorithms. Some of that is already happening, but it is clear that people don’t want more content. They already feel overloaded,” the report said, adding: 

“Advances in AI come at a time when human connection seems more important than ever in terms of other trends highlighted in this report such as personality-led news. The task for publishers is how to adjust to these new realities, to embrace technology where it makes sense while keeping humans in the loop — and to make the news more engaging and personal without losing sight of the values that attract people to their brands in the first place.”

Worthwhile links 

• GenAI and the intention economy: Your motivations are the currency of the future.
• GenAI and lawsuits: Looks like Anthropic scored a win — and a loss — in an intellectual property lawsuit.
• GenAI and lawsuits II: Hollywood joins in.
• GenAI and summaries: Lessons from The Wall Street Journal, Bloomberg, and Yahoo News on bulletpoint summaries.
• GenAI and summaries II: Wikipedia will no longer display AI-generated summaries on articles because human editors pushed back.
• GenAI and advertising: Reddit just announced new ways to mine its user-generated content for advertising insights, campaign planning, and product development.
• GenAI and advertising II: This ad was 100% AI-generated. Turnaround time: 48 hours. Here’s how.
• GenAI and chatbots: The Economist introduces SCOTUSbot.
• GenAI and privacy: LLMs can infer a lot about you.
• GenAI and reasoning: Models collapse at a high level of complexity, suggesting artificial general intelligence is still a long way off. 
• An AI diversion: UK court warns lawyers to stop using GenAI to produce legal arguments and witness statements because it makes stuff up and raises concerns about their competence.

About this newsletter

Today’s newsletter is written by Sonali Verma, based in Toronto, and lead for the INMA Generative AI Initiative. Sonali will share research, case studies, and thought leadership on the topic of generative AI and how it relates to all areas of news media.

This newsletter is a public face of the Generative AI Initiative by INMA, outlined here. E-mail Sonali at sonali.verma@inma.org or connect with her on INMA’s Slack channel with thoughts, suggestions, and questions.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .