The digital media space is changing faster than ever. To keep up, I’m launching a new monthly session with my good friend, Jez Walters from What’s New in Publishing. We’re going to sort through the noise and highlight the latest stories and developments that actually affect you as a publisher.
Here’s the rundown from our first session:
The AI Scraping Loophole: How They’re Getting Your Content
There’s a major disconnect between what AI companies say and what they do.
On stage at the recent FIPP World Media Congress, OpenAI’s head of content claimed they do not scrape through paywalls and that they respect robots.txt files. Yet, publishers know their content is showing up in AI models.
So, what’s happening?
A recent exposé from The Atlantic shed light on the loophole: Common Crawl. AI companies are going to this massive internet library—a repository of pretty much every URL known to man—and scraping publisher content from that source.
The takeaway: AI companies are honoring your robots.txt on your live site but appear to be getting your content from a backdoor like Common Crawl. This means you may need to investigate how to block your content from being indexed by these archives in the first place.
Is AI Search Killing Traffic? It’s Nuanced.
We’ve all seen the reports that AI-generated search summaries (like Google’s Gemini) are causing a reduction in traffic to websites. However, this doesn’t impact every market equally.
Who is getting hurt?
The large national and international news publishers. Their content is often a commodity, competing with dozens if not hundreds of other sources, and they are highly dependent on SEO.
Who is doing fine (or even growing)?
Niche and local publishers.
- City & regional lifestyle magazines like Boston Magazine, D Magazine, and 5280 in Denver are not seeing as much of an impact as national news. Their local information isn’t a commodity.
- B2B trade / association publications (that take to plumbers, restaurant owners, engineers, etc.) are seeing less of an impact because their information is highly specialized.
- Local business publishers like BizTimes Milwaukee and Ottawa Business Journal are actually seeing some growth in their SEO traffic.
This was even confirmed by an attendee during the live session, who noted that their small, regional, and niche magazines have seen better search traffic in 2025 than ever before. The more unique and specialized your content, the less replaceable it is by an AI summary.
The Untapped Goldmine in Your Archives
Publishers are obsessed with the “here and now,” but as Jez pointed out, the value in your archives is often massively underestimated.
He shared the case study of Autocar, a UK-based magazine that started in 1895. They undertook the “heavy lift” of digitizing all 6,600 of their back issues, which took six months.
The results were staggering:
- A 21-times uplift in views to their archive.
- An increase in present-day subscription sales as a direct result.
- The most popular content? Issues from the 1960s, 70s, and 80s, driven by what appears to be a powerful nostalgia factor.
Beyond subscriptions, there’s a growing revenue stream in licensing these digital archives to libraries and institutions using a “perpetual access licensing model”. This is a fantastic, recurring revenue source that many publishers haven’t explored.
A Critical Warning on Content Syndication
Before you rush to syndicate your content, you must read the fine print.
Many publishers syndicate their content through platforms like SmartNews and NewsBreak to get extra ad revenue and exposure. The problem is that if you read their terms and conditions, you may be giving them the ability to sub-license your content to anyone they want.
While there is no direct proof they are sub-licensing to AI companies, the clause gives them the right to. Be extremely careful who you’re giving licensing rights to, or you may find your content (and your archive) has been sold to AI models without your knowledge.
The Cookie That Won’t Die: Ad Execs Were Right
Remember the panic over the “sunsetting” of third-party cookies? Jez shared a fascinating story from Cannes Lion in 2021.
All the big media players (like Condé Nast, Hurst, etc.) were taking the end of the third-party cookie “very, very seriously” and diverting huge funds to build first-party data solutions.
But all of the ad execs in private said that Google would never go through with it. Google was making too many billions to kill the cookie and would string it out until regulators ran out of steam.
It turns out the ad execs were right.
Last month, Google officially killed its own Privacy Sandbox initiative—the technology meant to replace the cookie—and has allowed third-party cookies to continue.
While this is good news for programmatic ad revenue in the short term, it doesn’t change publishers’ long-term strategy. It’s still important to build robust first-party data
Quick Hits: Distribution & Deliverability
Finally, here are two more key developments you need to have on your radar:
- Nextdoor News – If you are a local publisher (city magazine, local business journal, parenting pub, etc.), you should by syndicating all of your content through a Nextdoor News Account. You have full control of the content you send to Nextdoor, their terms are very publisher friendly, and they link back to your site instead of repackaging your content within their app. Some publishers now see 10% or more of their total traffic coming from Nextdoor.
- Reddit Publisher Playbook – Reddit has released a new publisher playbook specifically to help media companies learn how to engage with its communities and drive traffic from the platform.
- DMARC Enforcement – All the major inbox providers (Gmail, Microsoft, Apple) are tightening their deliverability requirements. We firmly believe that in the first half of 2026, you will be required to move from basic DMARC monitoring to full DMARC enforcement. This tells providers to flat-out reject or quarantine (move to spam) any email that isn’t perfectly authenticated to your domain.