It’s becoming more and more apparent that AI is changing a vast range of things for us news media. At Google IO, multimodality was mentioned numerous times (almost as many as AI!). Very soon we’ll be in a world where we can switch modes at the touch of a button, click of a mouse, or a voice command.  

What does that mean for our UX?

I’ve been fascinated by the questions of how we help users navigate so much different content in so many different formats without creating a giant UX mess that makes navigation even harder. So today I want to share two conversations I’ve had recently.

Drop me a note if there is anything you’d like to see in this newsletter — whether it’s on a particular subject or a feature you’d like to see. I’m at Jodie.hopperton@INMA.org.

Examples of AI being used for UX in news media

The News Product Alliance ran a Webinar recently with Nikita Roy of Newsroom Robots on the AI of UX. Nikita gave some great examples of AI and UX, so I wanted to share a few of these with you:

• Politico runs AI summaries of legislation documents, which makes it easier for people to understand (for Pro subscribers only).

• YouTube summarises and groups comments; see below.

• Ippen allows you to choose how you want to read, either a long article or click on a button to read the summary.  This reminds me of Artifact doing different summarisations last year.

• Der Spiegel has a browser plugin that allows readers to give more context to a word or phrase, which is derived from the Der Spiegel archive.

• The Newsroom App uses AI to give you the top five news events based on what people are talking about. It gives different sources including progressive, moderate, and conservative.  

• Chatbots are, of course, a huge change in UX, which we covered in depth here (what to think about, how to build, data loop, cost, and RoI) and here (chatbots and users).

As I’ve written before, we may need to go further when thinking about UX, particularly as we see audio and video becoming more prominent. We’re already seeing changes with Google and Apple’s announcements, which may affect how consumers reach us. 

So much is changing right now that we don’t know the answers, but there is a lot of excellent experimentation going on. If you are doing something original with your UX, we’d love to hear about it. You can reach me at Jodie.hopperton@INMA.org.

Date for the diary: INMA Los Angeles study tour, October 21-25

If you are interested in how AI is meeting entertainment, come to do a deep dive into audio, video, AI, and commerce with me for a week in Los Angeles. We’ve just announced the first set of meetings and they are big: YouTube, Channel 1, KCRW, Verizon Innovation Lab, Bill Gross’s Idea Lab (and their soon to be announced AI for media initiative). Check out the INMA Los Angeles Tech Innovatoin Study tour here.

Q&A with Barak Ronen: AI for product is (almost) all about the user interface 

One of the areas I’ve been slightly obsessed with is how AI can (will) change our UX. We have so many different ways to tell stories and show information to our users, but it’s almost too much.

So I spoke to Barak Ronen, co-founder/CEO at the London-based software company Crux Knowledge, who has spent the last few years working to help news publishers drive engagement and habit formation through AI powered gamification and recommendation (more about how he was doing that here).

What led you to look at gamifying news and using AI? 

I was there for the whole print-to-digital, to social-to-mobile journey — and developed both a sense of an immense opportunity and an unease about how the industry wasn’t capitalising on this opportunity.

For me, we need to speed up how we reinvent the most important, money-making, audience-touching part of our business: the user experience. This has never been more true than now because this what GenAI primarily is — a revolution in how content can be experienced. 

What does that mean for a news organisation today?

AI can do (too) many things. The way to choose what to go for should be based on business objectives. What are we trying to optimise or drive? We need to stick to our business “whys.” Is it engagement? Retention? New subscriber acquisition? Time on site? Whatever the answer is, the feature that can drive that is what we should be building. Most of these metrics will be driven by changes to our UX. AI — and tech in general — is a means, not an end.

Where do you see tangible opportunities around new AI features that drive some of these quantifiable business objectives? 

A few areas are very interesting for me, and I’m intentionally skipping summaries and search here (not sure at all these are the biggest opps):

• Multimodal content consumption. Sites and apps are still dominated by pages that in many ways are similar to what we had 10 and 15 years ago. Text leading, video sometimes, a pic or two. AI can help us make bolder choices based on context and also open up new modals altogether. I can’t stop voice-talking to my ChatGPT. I’d love to talk to my favourite publisher or story that way. 

• Personalisation and onboarding. AI is a fantastic conversationalist. That’s what it was created to do. There’s a huge space of opportunity here to get AI talking to users about their preferences, their choices, to befriend them, and drive loyalty — and I dare say love. Imagine stopping guessing what users want — and them starting to tell us and enjoy doing so.   

• Avatars. Some of the best, sharpest, wittiest writers and opinionators work their magic in our newsrooms. They are our very own celebs. Their words matter not only because they contain information but also because of their voice and style. How do we create a system of “AI avatars” that leverages that, incentivises these newsroom stars, preserves truth, and builds their personal brands and the publisher’s brand? 

• Meta experience. When our audiences spend time with our content and interfaces, they are not only consuming information, they are also building their own identity and self-worth. They are doing something for themselves. AI can tell them what they are doing, give them insight into what they care about, what they know, who they are, and who their peers are — and it can do it in a way that is super engaging and brand building. In CRUX, we gamified news consumption in a Fitbit-style experience, but GenAI can take it to a whole new level.

What’s the biggest risk around AI’s impact on the user experience?

For me, the biggest risk is missing out on a UX that can actually feel better than the industry’s basic atom, the “story.”

Search disrupted the business model, social created a whole other content experience that ate up more of the revenue pie — but both of these did ultimately point back (respectfully?) to a well-written story. 

GenAI has the potential to relegate the “story” to a place of a footnote, a procedural evidence piece inside a GenAI conversation in someone else’s shop. I understand why publishers are making licensing deals with big AI tech, but we must think aggressively, too. How do we use the revenue from these deals to reinvent our own UXs? How do we use those footnote-style links to take users to an experience that’s more engaging than the age-old article page?

What is the thing you are most excited about in terms of AI and UX?

The complete reimagining of the front-page experience. The industry’s most iconic user flow is a person getting to the front page with an inquisitive mindset and a question: What do I need to know? 

Front pages deliver back through curation, hierarchy, design, and wording. They still look, broadly, not too dissimilar to print front pages. Zoom out and give it a few years — or months for those who dare. AI will allow us to deliver all of these things in a completely different way. I can’t wait to see that happen.   

How AI aggregate search looks: case study of Arc browser

Yes it’s here: New search experiences are coming out left, right, and center. They are AI led and AI summarised. It seems that even the aggregators are getting aggregated. 

On Monday, June 24, I upgraded my Arc browser on my phone (which is now my main browser), now called “Arc Search,” and was promoted to try the new search. Hearteningly, the top suggestion was about news (I am not sure if this is personalised; let’s assume not for now). 

Naturally, I clicked to see what I was served. A screen appeared noting a few sites that are being searched. In this came a few sources: Google, NPR, and Reuters show on the main screen but more are available upon a swipe. I find it curious that Google is also shown as a search: Are the aggregates being aggregated?

Once the search comes up (just a few seconds), the sources stay at the top and each story is summarised underneath. Scrolling, I got nine summaries total, each showing date, location, event, plus other categories that interchange based on the story contents such as participants, victims, impact, focus. It feels very Axios like.

At the end of the scroll, there is a section called “Dive Deeper,” which summarises the day’s summaries from various news sources. Note these aren’t based on individual stories but on daily round-ups of their top headlines, latest news, or day summaries. 

There is also a strong push to share. This share screen pushed up a number of times during the process.

What does this mean for news organisations? 

We aren’t sure yet, but the INMA team is constantly looking at different sources to unpack this move to GenAI search. We know some publishers are already experiencing double-digit drops in search, some over 30% (see Press Gazette reports here). We’ll be looking at different ways news organisations are compensating or planning to compensate. If you have views or work to share, we would love to hear from you.

About this newsletter 

Today’s newsletter is written by Jodie Hopperton, based in Los Angeles and lead for the INMA Product and Tech Initiative. Jodie will share research, case studies, and thought leadership on the topic of global news media product.

This newsletter is a public face of the Product and Tech Initiative by INMA, outlined here. E-mail Jodie at jodie.hopperton@inma.org with thoughts, suggestions, and questions. Sign up to our Slack channel.

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