SODP Dispatch - 28 May 2026

Nearly everything we use online is owned by big tech. There’s a better way forward

By Ashwin Nagappa

Globally, users of digital media are increasingly locked into a handful of operating systems, app stores, and communication platforms. Most of us must choose between Apple, Windows, or Android. All of these are owned by American tech giants.

Much of private and government IT infrastructure – websites, mobile banking, nearly anything online you can think of – uses cloud services, such as Amazon Web Services, Cloudflare or Microsoft Azure. They might have locations worldwide, but these are also US companies.

Mobile phones, laptops, smartwatches and more are mostly made by American or Chinese companies. And it’s getting worse as tech companies embed artificial intelligence (AI) assistants directly into everyday devices, such as Google’s Gemini or Microsoft’s Copilot. They’re doing this in ways designed to further entrench users within particular ecosystems.

When a single cyber security update brought down Windows computers the world over in 2024, it was a stark reminder nobody should put all their IT eggs in one basket.

But what might that actually look like? The “digital sovereignty” movement in the European Union (EU) can show us the way. European countries are gradually breaking up with American tech giants and pushing for local AI development, all in the name of achieving digital autonomy.

📊 SODP London Dinner — Publishing and AI

We are hosting a private dinner in London for senior leaders across academic, educational and trade publishing. This dinner is for senior publishing professionals who are actively thinking about how to keep their content economically relevant as AI changes everything. If you are making decisions around content licensing, AI integration, or new revenue models, this evening is designed for you.

Date: Tuesday, June 30, 2026

Time: 6:30 PM

Venue: 27c Eccleston Pl, London SW1W 9NF, United Kingdom

What you will walk away with

• Clarity on how publishers are approaching content licensing in the AI value chain

• Peer insight from senior leaders across publishing markets on what is working and what is not

• A practical understanding of new revenue models being built on top of publisher content

• Connections with a small, curated group of decision-makers you will not meet at a generic conference

The dinner is designed to be high-trust, collaborative, and off-the-record, with a focus on sharing what’s working and what’s not across markets. The evening will also feature a speaker session from a senior publishing industry voice, with more details to be announced soon.

A big thank you to Impelsys for co-hosting this evening with us.

Reserve your seat here ▸

📊 SODP San Francisco Study Tour 2026 – Publisher Ad Revenue Leaders

We're bringing together 8–12 senior publisher ad revenue leaders in San Francisco for a private, two-day programme on what actually drives signal and revenue in publishing today.

Hosted in collaboration with Ezoic, the programme takes place across partner offices in the city, a closed-door, operator-led experience built around one question: what is actually driving signal and revenue in publishing today, and how are publishers monetising first-party data in the AI era?

This is not a conference. No panels, no presentations, no vendor pitches, just candid peer exchange across working sessions, roundtables, and a private dinner under Chatham House Rules.

Topics on the table: audience signal vs noise, AI-driven discovery, commerce media monetisation, first-party data strategies, and infrastructure as a revenue driver.

Date: June 15 & 16, 2026

The speaker line-up includes:

• Nealesh Patel, Chief Revenue Officer at Crunchbase, on scaling publisher revenue from $10M to $100M+ and AI-era GTM strategy

• Joe Hyrkin, Former CEO of Issuu (Now a renowned CEO coach and tech investor), on revenue strategies, platform shifts, and what actually drove growth

• Amanda Hesser, Co-Founder & Chairman of Food52, on the economics of content, community, and commerce at scale

The programme also includes a Bridge-to-Bridge Bay Cruise, a private dinner at Taksim, and more.

Seats are getting filled. It’s an experience you don’t want to miss!

Request your invitation here ▸

🎯 Google Discover Content Audit

Most publishers are guessing why their content doesn't appear in Google Discover. They're optimizing blindly without understanding the actual pipeline their articles need to pass through. This free audit tool from State of Digital Publishing tackles that gap by mapping your content against the 6-stage Google Discover pipeline: ingestion and entity extraction, OG tag configuration, content classification (evergreen vs breaking), quality signals and E-E-A-T assessment, predicted click-through rate modeling, and user topic matching. It shows you exactly where your content is failing to qualify, whether it's poor image optimization, missing entity signals, clickbait patterns being flagged, or weak topical authority, the same diagnostic framework that determines which articles actually surface in users' feeds. If your team is publishing content hoping it reaches Discover without understanding the pipeline, give it a try. We built this to help publishers see what Google actually evaluates. We'd love your feedback.

Run the audit ▸

By Vahe Arabian

📄 Why llms.txt Is Just the Beginning and What Publishers and Retailers Actually Need to Survive the Agent Web

llms.txt adoption is spiking. HTTP Archive data now tracks it as a dedicated SEO metric, and the 2025 Web Almanac recorded valid llms.txt files on over 2% of desktop and mobile sites, notable for a standard that barely existed two years ago. Anyone watching the e-commerce and publisher space can see why momentum is building.

When Google announced WebMCP in February, a structured protocol for AI agents to interact directly with websites, handling product search, configuration, and checkout, the industry's hunger for a fast, practical response was immediate. llms.txt became the low-friction option. Quick to deploy, accessible without deep engineering resources. A signal, however imperfect, that your site is thinking about AI legibility.

But here is where I want to slow the conversation down.

llms.txt was designed to tell large language models what your content is about. It was never designed to make your site executable by agents. That distinction matters enormously right now, and I am not sure enough people are treating it that way.

WebMCP's proposition is different in kind, not just degree. It introduces structured Declarative and Imperative APIs that let AI agents perform real actions, not just read content, such as navigating checkout, configuring products and filing support tickets, without touching your visual interface. The analogy I keep returning to: llms.txt is a brochure. WebMCP is a set of keys.

The bottleneck I am watching is not technical. It is organisational.

Publishers and retailers who rushed the llms.txt deployment in reaction to WebMCP are solving a visibility problem when the actual problem is operability. Agents that cannot complete a task on your site will route around it to a competitor's site they can operate on. Content legibility buys you consideration. Structural operability is what converts.

The e-commerce implications are significant:

• Product discovery via agents requires structured, machine-readable attributes, not just a text summary of your catalogue

• Checkout operability requires stable, predictable form structures that agents can traverse reliably

• Trust verification between agent and site is still largely unsolved; authentication, consent, and tool-call security remain unresolved areas across the agentic web standards landscape.

Over the years, the pattern has been consistent: the first wave deploys the easiest available signal. The second wave builds the right infrastructure. The gap between those two waves is where competitive advantage accumulates.
llms.txt adoption is not the problem. Mistaking it for the solution is.

For digital publishers and ecommerce operators, where does your current agentic readiness actually sit? Content legibility or structural operability?

News companies now have 3 audiences — how best to serve them? | INMA

Until a year or so ago, most media companies believed they understood their audience. We knew there were different groups inside that audience. We built subscriber personas. We debated loyal versus casual readers. We invested in understanding formats, platforms, habits, and needs. Product teams focused on engagement. Editorial teams focused on storytelling. Marketing teams focused on conversion. And then came the bots. At first, they arrived quietly. Most publishers barely noticed them.

Read more ▸

You couldn’t create a more anti-news internet if you tried | NiemanLab

If there were a dictator of the internet who intentionally set out to destroy your ability to get accurate information, the result would look a lot like what’s already on your screen. But why? I mentioned here a couple of weeks ago that I’ve been studying economics to find more rigorous frameworks to describe why “creative destruction” has been better at destroying than recreating the news industry. The decline of original news by traditional media has not nearly been offset by the rise of newer media, mostly to the detriment of our democratic societies.

Read more ▸

Editorial cartoons find new digital life: Sack’s comeback highlights shift from newsrooms | Editor & Publisher Magazine

Journalists are used to overcoming adversity, but what if you’re a Pulitzer Prize-winning cartoonist and you suddenly can no longer hold a pencil? That’s the situation Steve Sack faced in 2022. The resident cartoonist for the Minneapolis Star-Tribune since 1981, Sack was a few years removed from defeating cancer with a combination of chemo and radiation when the fingers on his right hand began to feel numb. Four months later, his right hand was so numb he was no longer able to grab a pencil, pen or anything else he could use to draw his cartoons.

Read more ▸

How to embrace Nextdoor for News to grow local reach | RJI

Over the past year, Missouri Business Alert has experimented with Nextdoor for News as a low-maintenance way to expand our audience and connect with readers in our local communities across Missouri. For smaller newsrooms with limited staffing and social media bandwidth, Nextdoor offers an interesting opportunity: a largely untapped platform focused on neighborhoods and community conversation. While it’s not a replacement for owned audiences like newsletters or direct website traffic, we’ve found it can be a useful top-of-funnel awareness tool when used strategically.

Read more ▸

DuckDuckGo installs are up 30% as users reject being ‘force-fed’ Google’s AI Search | TechCrunch

Last week, after Google announced its huge overhaul to Search, I overheard a woman on the phone saying she was switching to DuckDuckGo because you can “opt out of using AI.” “Google just isn’t Google anymore,” she said. It seems that others had the same idea. At I/O, Google’s annual developer conference, the company said it would transform its search box into a conversational engine that expands for longer queries, anticipates user intent, and autocompletes searches.

Read more ▸

You’re Using AI at the Execution Layer. The Value Is in the Judgment Layer. | Duane Forrester Decodes

The tools are deployed. The licenses are paid. And if you’re a senior SEO or GEO practitioner right now, you’re probably using AI every day - for drafts, for summaries, for first passes at content that used to take twice as long. That’s real productivity, and it’s not nothing. It’s also not the return the investment is capable of producing. And the gap between what you’re getting and what’s available isn’t a tool problem. It’s a mode problem.

Read more ▸