📥 Hello, and greetings from the Central Office!
Alongside support and development work, our operations team completed several large maintenance tasks to keep systems running smoothly.
One major effort involved reviewing subscription records after migrating away from SendOwl.
Operational progress included:
• Reviewing 1,471 SendOwl subscription records
• Cross checking Stripe blocked and incomplete payments
• Ensuring unpaid licenses are revoked correctly
• Consolidating subscription data across systems
We also improved our internal support workflow by updating ticket statuses so engineering related issues remain visible and are easier to track.
These operational improvements help ensure better support response times and more reliable subscription management.
The Protection Service Built Its Own Threat
Cloudflare released a crawl API that scrapes entire websites with one request.
Input one URL, receive full site content in HTML, Markdown, or structured JSON.
Publishers who relied on Cloudflare to block scrapers discovered they couldn't block this new tool. Independent publisher Thomas Baekdal posted on LinkedIn that Cloudflare had "betrayed every single publisher."
Cloudflare's senior director of product James Smith told Digiday the company "probably didn't get this launch right" and apologized. The blocking issue resolved after Baekdal retested on March 16, 2026.
Six Surfaces, Zero Alignment
Cloudflare operates six different surfaces where publishers express bot-blocking preferences. Smith acknowledged the user experience "isn't where we want it to be" and promised fixes "in the coming days."
Different teams own different products. Fast-moving development meant launches didn't align. The company operates like a startup, Smith explained. Speed created the problem. The structural conflict remains. One company controls both defense infrastructure and crawling tools. Every product decision reveals whose priorities win.
The Intermediary Pitch
Cloudflare positions the crawler as creating a licensed AI content market. Smith told Digiday the tool "respects publishers' wishes" and existing controls. The pitch: give publishers more control, reduce inefficient site crawls, build compliant infrastructure. The company sits between publishers and AI firms. The crawler respects controls.
The messaging assumes publishers want licensing deals negotiated through intermediaries rather than direct relationships with AI companies.
Owned Infrastructure Eliminates the Conflict
WordPress sites on owned infrastructure face none of these structural problems.
You control the server. You write the robots.txt rules. You decide which crawlers access content and under what terms. No intermediary can launch competing products that undermine your protection strategy.
When AI companies want your content, you negotiate directly. You set the rates. You own the relationship. You capture the full value without platform cuts or intermediary fees. The infrastructure serves your interests exclusively because you own it.
Every Intermediary Serves Itself Eventually
Reddit signed $203 million in data licensing deals, according to its IPO prospectus. Publishers using intermediaries watch those deals happen without participation.
The company protecting you from scrapers builds its own scraper. The platform promising control changes the rules mid-game. The infrastructure you don't own serves priorities that aren't yours.
Cloudflare proved the pattern repeats. Owned infrastructure means you're the one signing the contracts, setting the terms, capturing the value.
What percentage of your content protection strategy depends on companies that could launch competing products tomorrow?
This week's radar reveals the infrastructure gap widening between platform-backed publishers and independent creators. From Digg's $20M collapse to Meta's piracy defense, the message is clear: ownership matters more than ever.
Digg's $20M relaunch dies in 60 days. Digg raised $15-20 million and died in two months. Bots overwhelmed AI moderation within hours of launch. CEO Justin Mezzell: "Trust is the product." Platform infrastructure collapsed when gatekeeping depended on AI instead of ownership models that creators control.
AI pays Disney millions, scrapes creators free. AI companies claim fair use while signing multimillion-dollar deals with Disney, Condé Nast, Vox, Warner Music. Patreon CEO Jack Conte: "If it's legal to just use it, why pay?" Patreon positions as licensing intermediary for hundreds of thousands of creators locked out of direct negotiations.
BuzzFeed down 17% as AI splits by size. BuzzFeed revenue dropped 17% year-over-year to $189.9 million in 2024. Large publishers negotiate AI licensing directly. Small publishers get locked out entirely. Arena Group now employs full-time staff navigating AI content marketplaces-a role that didn't exist 18 months ago.
Talent agency goes direct to theaters. Night Media executive produced theatrical film with Theo Von (20 million monthly YouTube views) and David Spade. Self-distributed to Cinemark and Regal April 17. Agency bought podcast network from Warner Bros. Discovery after Rooster Teeth shutdown. Vertical integration from audience ownership to theatrical distribution.
Meta claims piracy redistribution is fair use. Meta downloaded pirated works from Anna's Archive via BitTorrent. Now claims redistributing those works qualifies as fair use because distribution is "part-and-parcel" of BitTorrent technology. Legal argument: simultaneous upload during download doesn't count as distribution. If this holds, every scraping operation can claim technical necessity as legal defense.
The new flow: User → AI model → Search engine → Website → AI synthesis → User. You lost direct access to 60% of the discovery process.
Yoast's Principal Architect Alain Schlesser maps how AI systems now function as gatekeepers between users and websites.
WordPress site owners can optimize schema markup and structured data to influence how AI models retrieve and synthesize their content during this intermediary step. Platform creators have zero control over how TikTok, Medium, or Substack structures their content for AI consumption.
That’s all for this week!
Michael - Operator @WP Folio - now WP Defense Lab. Same Plugins. Different Name.