Hi all. This month we’ve got pieces on Substack, Q&A with Ethan Mollick on AI in organizations, a podcast on how metrics make us miserable, plus my column on the view of media from the Game Developers Conference and South by Southwest.

Both Amber and I will be at the PBS Annual Meeting and pre-conference Technology Summit in Austin next month. If you’ll also be there and are interested in chatting about AI, games, innovation, or just the future of public media look us up! You can also drop me or Amber a line ([email protected] and/or [email protected]) if you want to book a specific time to chat.

Webinars That Track

There’s still time to register for today’s April webinar at 1pET/10aPT. We’re collaborating again with Current to bring you What’s Working with Public Media Newsletters. Email newsletters have become a powerful tool for building loyal audiences and creating a direct line to the communities we serve. Unlike social platforms, your newsletter list belongs to you — no algorithm, no platform risk, just a direct connection to the people who want to hear from you. In this webinar, you colleagues from across public media will share what they've learned building newsletters that actually work, from reimagined flagship products to multi-newsletter portfolios and experimental formats. Each speaker will walk through their work in a show-and-tell format, giving you a look at real newsletters in the wild and the thinking behind them. Whether you're launching your first newsletter or rethinking an existing one, you'll walk away with practical strategies and fresh ideas to take back to your organization. Register here!

I haven’t done a newsletter edition since our March webinar on working with creators was ready to stream. So, if you want to revisit or check out Myths & Truths of Working with Influencers, here you go:

And if you’re interested in this topic, you’ll also want to review (or revisit) the American Press Institute’s guide to influencer collaborations, published last fall.

Last month I spent about 10 days on the road at the Game Developers Conference in San Francisco, followed by South by Southwest in Austin (where, ICYMI, Paula Kerger had a keynote). What follows are some reflections based on what I heard from ‘the innovati’ at those conferences.

As the fog of hype around AI burns off and its true utility comes into focus, one demonstrable advantage of generative AI tools to creatives is the ability to show instead of tell. Call it "demos over memos" (or, if you prefer alliteration to rhyming, "demos over docs"). Whether it’s rapid ideation and iteration, or a quick 80%-complete prototype, AI allows us to combine craft and concept further up the ideation value chain and communicate our intent for a finished product, story, or experience.

Intent is perhaps the biggest ingredient of the creative process that cannot be outbrained to AI. AI can get you to "average" fairly quick. That’s a good thing. But AI does not remove the necessity of thinking through that problem space of creation. Good ideas, good stories, good products, are still hard. Addressing that challenge requires intent. As Irena Pereira, Founder & CEO of Infinite Realms/Unleashed Games, put it at GDC, “AI is the forge, not the fire.”

For public media creators that forge means that the scope of possibility on human creative intent is wider than ever. It also means that creation is no longer limited to the speed of human thought. Across both conferences, there were numerous references to AI tools as a "partner," "participant," or "collaborator" with whom modern knowledge workers "share," replacing what Jon Gibson and Tara Phillips of Keywords Studios call a “prompt and pray” approach to AI with “steer and iterate” interactions. I also heard the metaphor of orchestration used more than a few times, usually associated with the idea that knowledge workers and creators are constantly tuning the system while moving from "doing to deciding."

Again, this is a good thing. But there is a creative responsibility that comes with the opportunities afforded to us by AI: creators must lean into the friction of craft.

Orson Welles once said, “The enemy of art is the absence of limitations” (which I misquote all the time saying, “Great art through limitations”). And especially at GDC, there was a lot of consideration of friction as a necessary, constructive component of craft.

Concepts plus craft can be multiplied by constraint, and that equation can yield something that meets our standards in public media. Workflows should be crucibles. But AI tools offer creators a tempting lack of restraint, and that lack of restraint is one of the key ingredients of slop.

Every month someone is releasing a better AI model or product with novel features. As Aparna Chennapragada said during her SXSW session, “Today's magic is tomorrow's commodity.” It’s important to remember that our measures of success for public media are not aligned with the success of AI. Novelty is the AI companies' stock in trade right now and while novelty in public media is nice it’s not core to our values.

What is core to our values is trust. And Mark Cuban, of all people, put forward a pithy equation on trust during a session he did on the future of journalism: Trust equals transparency of process divided by clarity of motives. Like intent, process sets us above a vast generative wasteland of "slop.”  To process is human, and it’s the process loop that matters most. In a world where the cost of information is going to zero, the value-add of public media comes from our creation (e.g., editorial) workflow.

That's the friction against which we need to optimize. Given the tools to which we have access here in mid-2026, the best way to do this is to identify pain points in your workflow and then consider how AI can help smooth those down to constrictive frictions.

Creatives, you’ll know you’ve found the right balance when the process leading to the pain points becomes “fun” again.

Public media leaders, you need to be setting your creation teams for success. Help your teams set principles for AI use, don’t hit them with mandates. When generation is cheap and getting cheaper, standards matter more. They are the ultimate friction against which we hone our value.

While we are leaning into the constructive frictions of craft, there is a destructive friction threatening our industry’s flank. This is the friction audiences experience in finding our content.

In an AI-mediated discovery environment, citation is not a vanity metric; it is survival. In our digital lives "discovery" is the whole ballgame; from finding photos on your mobile phone, to finding content to consume while walking through the airport, to assessing who to vote for in the next election. Increasingly, that information will be intermediated by AI. Everywhere I travel, I hear Gen-Z and Gen Alpha people talking about ‘ChatGPT'ing’ information.

To address this destructive friction, public media needs to prioritize a discovery strategy that addresses AI via SEO (or AEO, or GEO). The consensus from Austin and San Fran is that AI is currently in its listicle phase. This is a natural extension of a media model built around the probability of a click, and the more you can structure your information the better the chance it will re-surface with you cited as a source. At this point, citation is the goal.

We need to accept that this isn't aligned with storytelling for an audience of people. I'm not suggesting that we always subjugate storytelling to the bots, after all stories are largely the reason that any of us know what we know. But at this point AI isn’t prioritizing the unweaving of your carefully knitted narrative to answer a prompted question. We need to accept there are times we need to choose bot-friendly, choose human-friendly, or choose to structure the story for both (e.g., bulleted summaries of key points at the top of each story). The clock is ticking. In one workshop on AI search, the presenters predicted that AI search will overtake traditional "10 blue link" Google search by 2028.

The final takeaway from my marathon of professional development is that gaming is the future. I see this happening on two levels. First, for Gen-Z and Gen Alpha, gaming is (and will remain) a public square.

Second, if you want to see what is coming in the near future for journalism and media, look at what is happening economically and technologically to the gaming industry. This seems especially true when it comes to AI.

The games industry is an industry that acutely feels the threat from AI. It is an industry that places a high value on craft. And it is an industry that wrestles with discovery at an epic scale. What the games industry is figuring out now about AI, craft, labor, and discovery, the rest of media will be forced to learn soon enough.

Okay, on to the links.

Building the Public Infrastructure for Local News (Alex Frandsen - Governing)
- This is the line that really hooked me in this piece: "So saving the existing system isn’t the goal. Building a healthier one is. That requires lawmakers to start with a different set of questions: What civic information do residents actually need to stay safe, participate in democracy and build power with one another? And what kind of structural changes do we need to usher in the kind of local media system that’s capable of creating that information in abundance? This reframe shifts the center of gravity from industry salvation to community impact and positions the public as a key partner in debates about the future of local media, rather than as a passive bystander." They feel like the fundamental questions that we should be wrestling with at our local organizations.

CPB’s Demise Has Potential Impact On Commercial TV (Mary M. Collins - TV News Check) - I appreciate the rallying cry, even if it is wrapped in the candied shell of self-interest.

BBC will cut up to 2,000 jobs to reduce costs by about 10% (Brian Melley - AP) - As my mom is fond of saying, "Things are tough all over."
Related: And also in Germany: Budget reduction weakens Germany's international broadcaster

The $1 Million Gamble: Why We Threw Our Fundraising Goals Out the Window (Rob Risko) - Anyone who knows me knows that fundraising is not my métier. But I appreciate LAist showing their work. And I hope this inspires others to work within the unique characters of their local markets and regions to explore similar innovations in local support.

Guardian’s first Substack experiment is republishing food newsletter (Charlotte Tobitt - PressGazette) - Since the topic of this month's PMI webinar is newsletters, I thought I'd share this piece about how a major media org (and one not dissimilar to public media) is approaching Substack.

Substack, Polymarket announce partnership: ‘Journalism is better when it’s backed by live markets’ (Dominick Mastangelo - The Hill) - Is it? That's a question any of us with newsrooms will have to wrestle with at some point this year. As I've suggested before, the time to start developing policies around this is now. But that aside, if your newsletter strategy does or might include Substack, this is something to note.

AI Disclosure in Podcasting: A Decision Framework (Alberto Betella) - H/t to Collin Berke, NPM’s Director of Research, tipping me to this. Across your organization, people should be A) experimenting with AI and B) asking this question about transparency of use. I'm not necessarily advocating that you adopt this framework as it, but I think it could be a good jumping off point for setting your own disclosure policy at your organization.

How AI is transforming freelance journalism (Marina Adami - NiemanLab) - The use of AI is being normalized into our industry. This piece offers a good view from the trenches. You might find, or be inspired to create, some new workflow hacks here.

Q&A with Ethan Mollick (Tom Rachman - AI Policy Perspectives) - This is a great check-in with one of our favorite thought leaders on the practical use of AI. I’d especially recommend this one for GMs and department heads, as it speaks to leadership in organizational use of AI, and the need to rethink how we set up our organizations.

You Can't Influence What You Can't See: Why Your Organization Is Already Losing the AI Debate (Robert Bole) - While this short piece isn't specifically on public media, Bole worked at CPB 15 years ago and has been at the intersection of media, innovation and DC ever since. So, it's not just confirmation bias that makes me believe him when he says the information environment has changed.
Related: On a similar theme, check out Alex Curley's piece in Editor & Publisher: Newsrooms face AI control crossroads

Why AI Needs Us (Tim O'Reilly with Claude Opus 4.6) - There are three key parts to this piece. The first is the essay by Claude, which reads in the style of a profile that might have been in the New Yorker, or one of the other Condé Nast publications. The second is O'Reilly's commentary at the end on the process, because I think the process matters as much as the effect and O'Reilly has always been good about showing his work. The third is the effect, which is to say how you feel about this work upon reflection. Case in point, I felt uncomfortable with Claude's use of the first-person perspective. I can't deny that I liked some of the turns of phrase, yet they simultaneously resonated with me and felt out of tune. hollow and dissonant to me. But, that’s just me, and as we say here, your mileage may vary.
Related: This piece by Emily Bender and Nanna Inie from Tech Policy Press, is a nice chaser to the Opus O'Reilly piece: We Need to Talk About How We Talk About 'AI'
Also Related: Why comparisons between AI and human intelligence miss the point by Celeste Rodriguez Louro and Jennifer Rodger.

Introducing ChatGPT Images 2.0 (OpenAI) - Honestly, that OpenAI has (finally) stepped up its game in image generation again is almost secondary to how they present the announcement. And I'm not just referring to the demo video. Scroll down the page. It's clever, even if it sacrifices clarity of communication in the process.
Related: Here is one company snapshot analysis of this new tool, which is probably as interesting to me because of how they categorized the outputs they judged as much as the scoring itself.

Deep Research Max: a step change for autonomous research agents (Luka Haas & Srinivas Tadepalli - The Keyword) - I still quietly chuckle when I remember that just three years ago, we were adamantly advising against using AI for research. Recently we've been in the Wikipedia-replacement zone (i.e., start with a Deep Research report for high altitude context). But we're moving to something different. I can't imagine doing a strategic planning session without your strategy team actively engaging AI at every step of the process. At the content creation level, a tool like this one should be in your toolbox for any non-fiction content. Of course, you still have to fact check (so long as LLMs are the basis for the AI tools we use, I don't see that changing), and you still have to spend time reflecting on the research before turning it into context for our audiences.

Introducing ElevenLabs for Government (Oswin Kruger Ruiz, Ben Supple, Grace Smith - ElevenLabs) - This announce is a little dated, but since a lot of stations are tied to state or local governments, this is potentially one to check out.

Health information delivered as a video game can bridge the communication gap between patients and providers (Elena Bertozzi - The Conversation) - This is a great essay on the power of 'games for good.' At our core, humans are built to play; way more than we're built to doomscroll (or, if you're old enough, channel surf). At this one to you evidence box for why public media organizations have every reason to be in the business of making games.

How Metrics Make Us Miserable (Plain English with Derek Thompson (and C. Thi Nguyen) - I found this episode to be a bit of a two-fer. The title really only refers to the first half, but that part is good for thinking about the KPIs we've been trying to apply to public media for the better part of two decades, and how they may actually carry unintended incentives with them. There is some important reflecting to be done there. Optimization to KPIs doesn't necessarily make for great art. And public media has always existed at the intersection of art and product. Then, in the second, part the conversation turns toward the philosophy of games. What is a game, and why is the playing of games such a human endeavor?

Another h/t to Collin Berke at Nebraska Public Media for this one.