Three Stories I’d Like to See in 2026

Welcome to Second Rough Draft, a newsletter about journalism in our time, how it (often its business) is evolving, and the challenges it faces.

A new year, among other things, is a time when many reporters and editors try to set their agenda for the months ahead, and in this annual column I endeavor to help. Here are three (or four) big stories I’d love to see pursued in 2026:

A window into whether there’s an AI bubble?

There seems increasing recognition that our economic tale these days is one of two economies, the AI boom (bubble?) and the downbeat rest. I accept the consensus that AI will likely eventually be at least as revolutionary as was the consumer internet, but the key word in that notion is “eventually.” Whether we are at the beginning of a genuine boom or somewhere near the height of a bubble that may precede it (as occurred in 2000) will be one of the critical questions answered this year.

For business journalists, there is an important story here that I think has been insufficiently explored. The case for a bubble hinges, in significant part, on how much of the huge revenues of the leading AI companies are involved in “circular” activities, that is, cross-investments and paying each other for chips and related computing power and facilities, comprising internal growth and infrastructure buildout. Because many of the companies involved are still private, this is a harder question to answer than it might otherwise be.

 

What journalists could bring to the fore is the flip side of the issue: how much end user revenue does AI already command? The fear is that the answer is a very small fraction of all those tens and hundreds of billions. How could that be true? Well, think about your own AI usage. I employ these new tools frequently for research, and to create most of the illustrations for this newsletter. I have never paid a penny for any of that.

I am mostly retired, and AI is, at least so far, more of a business-to-business than business-to-consumer phenomenon, so let’s look also at our industry more broadly. I talked to one publishing executive recently whose fairly large newsroom uses or may soon use AI for research and data analysis, drafting article summaries and versioning, and headline, social media and other promotional content. Their business side employs it for drafting a fleet of customer-facing materials. But for a self-restraint currently common in newsrooms, this list could extend as well to support for graphics and illustrations.

For all of that, they aren’t yet a paying AI customer either, given that their lead technology provider, like most everyone’s, is also an AI platform. The newsroom’s AI capability, that is, is already bundled into their existing technology spending. The highest annual run rate of AI outlay the executive could currently imagine directly reaching in 2026, in a very aggressive scenario, amounts to just a mid-five-figure total, only a few hundred dollars per average staffer. Even if factoring in the increased license fees enterprise software vendors are seeking to cover AI enhancements, total spending stays in that same ballpark, at least in the foreseeable future.

Given those sorts of numbers for newsgathering and publishing, which is, of course, a very information-intensive industry, it is hard to see vast end user revenues almost anywhere right now. But reporting this out, and especially beginning to quantify it at scale, could be a story of enormous near-term consequence.

Restoration or reinvention in “public media”?

More parochially, 2026 will also be a year of decision in what used to be called “public media” or domestic public broadcasting. (Without public funding, it’s really now just an important sector within the larger world of nonprofit news.) The issue here is whether stations (and others) are going to focus their efforts on a restoration of the system as it existed before the congressional rescission, or instead are willing to attempt a reinvention of their business and operating models.

There will be stations and networks, and philanthropic funders, following each of these contrasting routes. I hope the journalists covering all this activity will work to assess the various initiatives in this context of restoration vs. reinvention. For myself, I continue to believe that reinvention is the only viable course in the longer run. The likely failure, this coming summer, to be able to repeat the flood of small contributions that responded to rescission should add emphasis to this argument.

What does Trump actually know?

Absent some deus ex machina, Donald Trump will almost surely be the dominant news generator of 2026. I think one of the few under-reported stories on his second term so far is the question of how much he believes the nonsense he often spouts. The reason this seems to me significant is not because of some personal affinity on my part for abnormal psychology, but because I think the answer could shed important light on how isolated Trump has become. Personal isolation in presidents, history teaches us, is almost inevitably dangerous.

So, for instance, when Trump says the price of gasoline is lower in many places than it is anywhere, or when he says inflation has been licked, or that manufacturing is reviving in the US, or that peace has broken out somewhere armed conflict continues, is he just trying to deceive people, or is he ignorant of the facts? There have been a few stories that tried to get at this; this Semafor piece is one. The New York Times on Trump’s second term work habits adds a bit. But there is much more such work to be done, and it will likely require one or more of the largest White House and political teams to do it. This piece last week, mostly on Trump’s health but also delving into his work habits, from the Wall Street Journal (which has done probing reporting on the Epstein scandal) struck me as a bit credulous.

I would love, for instance, to see a quantitative rather than anecdotal study of Trump’s social media posts from 2025 compared to those from 2017 that analyzes his sources of information. How have his viewing habits changed? Who is he newly tracking, and who has he stopped following? Beyond this, if Vanity Fair can play on egos to get the White House stories they published last month, can we get some folks in the West Wing to talk more about the paper flowing into Trump, and the paper that isn’t, about who can regularly get through, by phone or in person, and who cannot?

For those who have read this far, here’s a fourth story idea, admittedly probably harder to get: What is up with Trump’s health? With him rapidly visbly aging and tiring, at least one purported MRI (he and his doctor now tell the Journal it was actually a CT-scan, which found no cardiac issues) and many unexplained apparent IV bruises, it would be very disappointing if the press repeated its mistake with the previous president of failing to answer such obvious questions. I know I’ve said this before, but it bears repeating.

Those of us who still enjoy following the news feel that way, in large measure, because we are sure there will be unexpected developments this year, as there always are. And the solid reporting of day to day events will absorb much of newsroom energies. So the news agenda for 2026 will, again as always, be largely outside the control of reporters and editors. But I hope that, to the extent they remain the masters of their own time, some will take some of it to explore these key questions, and endeavor to get readers the answers they deserve.

Leave a comment