How’s your December going? It’s that time of year when my inbox is filled with predictions for the year ahead. Since GenAI is such an exciting, fast-moving space, I thought I would share some of the more interesting forecasts with you in this newsletter before we start winding down for the winter holidays.

The first batch of predictions come to you from some members of the INMA Generative AI Advisory Council, a group of brilliant AI practitioners from media companies in different corners of the world who meet once a month to discuss what they are doing, thinking, and preparing for. 

The second batch is a broader sweep of forecasts for technological change that will have an impact on how we do business. A common theme running through them? How we will need to consider the impact of GenAI on humans in 2025.

Enjoy!

Sonali

Note: I’m working on INMA’s GenAI programming for 2025. What would you like to see? What are you working on that is really interesting? Drop me a line and tell me so our Webinars and workshops can be valuable for our community! Don’t be shy.

Redefining the unique fingerprint of news brands 

Thinking beyond AI

Toughts from Lyn-Yi Chung, deputy chief editor at CNA Digital in Singapore:

To stand out, newsrooms will have to start thinking beyond AI for efficiency and optimisation. They should focus on creating intuitive, AI-powered experiences that give visitors a sense of time well spent on their site or with their brand.

Gone are the days where creating endless reams of content, hoping that something gets carried by the winds of third-party algorithms is a viable strategy. You don't want to be relegated to an occasional footnote on an AI answer engine.

Newsrooms will also have to work on dealing with testing fatigue for AI solutions among their handful of champions and early adopters, and formalise the setting up of cross-functional teams to spread the load.

Newsrooms that embrace product-thinking, strong user empathy and that can make AI more accessible across the board will likely race ahead.

How AI will evolve

Thoughts from Rohit Saran, managing editor, The Times of India in New Delhi:

At the individual level, I expect AI’s impact to move from solving “first-generation” issues (e.g. grammar, structure, translation, transcription …) to addressing “second-generation” issues, like being reporters’ and editors’ digital twin helping them do better and faster ideation, research, writing, coding, publication.

At the institutional level, in large newsrooms, I expect AI to start delivering measurable and system-wide improvements in workflow simplification, time and quality improvements, and product improvements. In effect, company-wide improvements in all 3 Ps — process, product and people. Definitive ROI on AI will be visible.

At the industry level, I think individual journalists and “non-journalist” storytellers will become a bigger and more formidable challenge to mainstream news media companies. This could lead to complete remaking of the news media business in two to three years.

A balancing act

Thoughts from Sarah Owsik, AI business partner at Mediahuis in Brussels:

2025 will be about smartly balancing between creating near-time value quickly and adapting and learning to navigate into the future. 

For 2025 and beyond, the key will be to strike the right balance between thinking aggressively about the big changes that will come and thinking small enough to move forward with creating near-time value.

This entails strengthening collaboration between journalists, AI experts, and product experts, defining together what “good” means in the context of GenAI as a starting point to learn and improve. The more we get GenAI solutions empowering our journalists integrated in their workflow in an intuitive way, the better we can accelerate adoption.

But when it comes to navigating to the future, this won’t suffice.

GenAI will keep on challenging us to sharpen what sets us apart as a publisher, to define and redefine what the unique fingerprint of our journalism is, the specific fingerprint GenAI doesn’t have. Although we don’t know where we will end up in the longer term, we’ll further shape a common language within our organisation when it comes to thinking about the future so we’ll know what to do next.

Looking outward, you can already see the potential paths to more and new value creation with GenAI in 2025. The rise of AI agents will translate to new possibilities in streamlining workflows and interacting with readers as the technology evolves. Also, Big Tech enabling their AI assistants to interact with other products — for example, Spotify joining the Google Gemini ecosystem gives food for thought.

Next year will be about generating value with GenAI in the way we bring our journalism in our products to our readers. But the keyword here is value, and not GenAI, keeping in mind that our readers define where true value lies.

While GenAI becomes more sophisticated, the human role will become even more critical. Our unique abilities — creativity, empathy, contextual understanding, and our moral compass — are essential in maintaining journalistic integrity and trust.

Second looks

Thoughts from David Cohn, senior director of Advance Local’s Alpha in Berkley, California:

The hype wave around AI has peaked (or close to it), but its impact on the industry hasn’t yet been felt. 

What I predict isn’t that AI will take us “boldly where no one has gone before.” Instead, it will help us revisit old territory, digging deeper into abandoned wells to uncover untapped value, hidden insights, and opportunities once deemed impractical. 

So, what are some “pivots” that were once deemed infeasible that might be worth a second look using new technology?

Human-centred interactions

Thoughts from Liz Lohn, director of product/AI at The Financial Times in London:

GenAI will drive a return to offline, experiential, and peer-led engagement. The oversaturation of digital channels and declining trust in algorithmically curated content are driving a profound shift in audience behaviour.

Instead of passively consuming endless feeds of personalised content, audiences are seeking richer, more meaningful ways to engage — through immersive experiences, offline interactions, and trusted micro-communities.

In-person and hybrid events will keep playing an increasingly vital role in this shift, becoming key environments for consuming information, finding inspiration, and building authentic connections. Platforms like Reddit and Substack, which prioritise strong communities and individual creators over traditional brand-led models, will continue to solidify their position in news consumption. 

For news organisations, this represents an opportunity to go beyond simply delivering content and instead foster human-centred interactions — whether by designing dynamic event experiences, nurturing niche community spaces, or amplifying the voices of trusted peers.

Agents will lead to significant internal change

Many business leaders are also thinking about the impact of AI on humans as we head into 2025.

“This is the year of AI utility,” said Sridhar Ramaswamy, CEO of data storage company Snowflake. “It’s a defining moment: Those who integrate AI thoughtfully will drive efficiencies that free up resources for critical tasks, making 2025 an ‘adapt or die’ year for organisations. AI can streamline workflows, unlock previously inaccessible insights, remove intermediaries, and more. Businesses that implement AI strategically will see a measurable impact in the hours of work reduced, millions of dollars saved, and accelerated pace of innovation.

“As a part of this transition to utility, AI has changed the game in that everyone is now a data user. Generative AI has made data and data analysis instantly accessible via conversational language, and that will change professional roles across the board.

“Business intelligence, once a specialised function, is now being democratised. Professionals from all backgrounds are tapping into data-driven insights, and as users gain comfort with AI, data roles will evolve. CEOs must anticipate this evolution, investing in upskilling initiatives and aligning roles with AI’s capabilities, ensuring that every employee is equipped to contribute to a data-centric organisation.”

Research group Gartner also expects the impact on staffing to be significant. It predicts agentic AI will be mainstream within the next two to three years, providing businesses with a digital workforce that doesn’t need vacations or benefits and can do things with or without us. These agents will be goal-driven, can make decisions, and will possess memory as well as the ability to plan and sense, and tools they can use.

Indeed, Deloitte Global expects a quarter of the enterprises using GenAI to deploy AI agents in 2025, growing to 50% by 2027.

“While the ultimate aim is to achieve autonomous and dependable agents, Deloitte expects significant improvements in their capabilities in 2025 as these technologies rapidly advance, with agentic AI moving past pilots and proofs of concepts in some markets and for some applications in 2025.”

This could make Web sites and apps unnecessary: “There are many things we go to Web sites for to get information and make a decision — we could have an agent do that,” said Gene Alvarez, distinguished vice president/analyst at Gartner.

It would also make it easier to upskill or onboard a workforce by creating a digital buddy to new employees or bring new skills to staff by letting an agent work as a co-worker that teaches as well as does. According to Alvarez, the question then becomes: How will we govern this new workforce of ours? 

Advances in GenAI will also mean more leisure time for workers, according to Zoom CEO Eric Yuan.

“Over time, AI digital assistants will become fully customised, evolving into a ‘digital twin,’ equipped with your work history and institutional knowledge. They will fundamentally change how we work, allowing us to prioritise the creative and strategic tasks that truly require a human touch and thoughtful focus time.

“Back in 1926, Henry Ford revolutionised the work week, reducing it from 48 to 40 hours. Now, thanks to AI, a four-day workweek could become the norm — enabling the fifth workday to be used as a mindful retreat from daily tasks to be spent with friends, family, or hobbies.

“In the end, our digital twins will allow us to top Ford — and the implications for productivity, work-life balance, and collaboration will be profound. A four-day work week could truly be within reach.”

Well, there’s a thought as we head into the holidays. Wishing all of you a relaxing break, and looking forward to hearing what you’re working on in 2025! 

Worthwhile links

• GenAI and lawsuits: Five Canadian news publishers — Torstar, Postmedia, The Globe and Mail, The Canadian Press, and CBC/Radio-Canada — are suing OpenAI.

• GenAI and sources: The Columbia Journalism Review points out that ChatGPT does not cite sources accurately, even for its partners:

• GenAI and partnerships: Looks like Amazon is looking to partner with and license content from news publishers for its new version of Alexa.

• GenAI and deals: Publishers are leaving money on the table with GenAI deals, such as the 50-50 revenue split with intermediaries such as Prorata.

• GenAI and deals II: Telecommunications company Orange will partner with OpenAI and Meta to develop AI models based on African languages.

• GenAI and training data: Harvard releases a massive free AI training data set to equalise opportunities among large and small AI companies.

• GenAI and writing: More than half the posts on LinkedIn are written with AI.

• GenAI and music: The Suno app produces music that is better than the output of 80% of the students at Berklee College of Music, their professor says.

• GenAI and tasks: 15 things AI is good at: excellent list, with brief explanations. 

• GenAI and banking: Scams are down 50%, and call-centre wait times are down 40% thanks to AI at Commonwealth Bank of Australia.

• GenAI and geopolitics: Russia teams up with China, India, Brazil and others in battle for global AI supremacy.

• A Christmas AI diversion: Ask and AI Jesus shall answer, but He does want to safeguard your privacy.

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.

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