WordCamp US 2026: What AI Means for Hosting and the Web
Phoenix showed the industry conversation has already shifted to AI, infrastructure and security

WordCamp US 2026 may have been a WordPress event, but many of our most important conversations in Phoenix were not really about WordPress at all. They were about AI, and more specifically, what AI is going to mean for hosting, cloud infrastructure, security, support, data centers, and the businesses built around them.
As we moved between sessions, dinners, hallway conversations, and meetings with attendees, that pattern kept repeating. We often found ourselves spending more time talking about AI than WordPress itself, and that was one of the most revealing takeaways from the entire event.
It reinforced exactly why AI needs to be at the center of the conversation at Atlas Digital Summit this September in Reykjavík.
The Conversation Has Already Changed
A few years ago, conversations at an event like WordCamp US might have centered around performance, plugins, hosting, WooCommerce, security, or the future of the CMS. Those topics are still important, but AI now cuts across nearly all of them.
We talked with people about AI-assisted development, automated support, infrastructure management, security threats, agent access, data center demand, and how customer expectations are changing. These were not isolated conversations with one or two people, they came up repeatedly across different parts of the ecosystem.
That is what made Phoenix so interesting. AI was not simply a keynote topic or something vendors were adding to product messaging. It had already become part of the everyday business conversation.
We Talked About AI More Than WordPress
That may have been the biggest surprise of WordCamp US.
We were surrounded by WordPress, yet conversation after conversation moved naturally toward AI. People wanted to know where it was going, how quickly it was changing, what risks were emerging, and what it meant for their companies.
The discussion was also becoming much more mature. It was not simply, “How do we add AI to our product?” Instead, people were asking how AI changes the hosting business, how it changes development teams, how it affects customer support, and what happens when agents begin performing tasks that previously required people or traditional software.
Those are no longer product questions. They are strategy questions.
AI Is Becoming an Infrastructure Issue
One of the strongest takeaways from Phoenix was that AI is quickly becoming much more than a software discussion. Once AI agents begin interacting with servers, databases, APIs, applications, and customer environments, the conversation becomes an infrastructure discussion.
That immediately creates new questions around access and control. What permissions should an agent have? How much autonomy is too much? What happens when an automated process makes the wrong decision, exposes a credential, changes a configuration, or acts on bad information?
There is also the security side of the equation. Attackers are gaining access to the same class of tools, which means vulnerability discovery, exploit development, phishing, credential attacks, and large-scale scanning can all become faster and more automated.
Hosting and cloud providers are going to have to think about both sides of that equation at the same time. AI can make infrastructure easier to operate, but it can also introduce entirely new ways for infrastructure to fail.
The Data Center Conversation Is Part of It Too
The AI discussion is also moving deeper into the physical infrastructure behind the internet.
AI requires compute, storage, networking, energy, and data centers. As AI workloads increase, questions around capacity, efficiency, sustainability, resilience, and where that infrastructure is located become more important.
That means the AI conversation cannot be limited to software companies or developers. Data center operators, cloud providers, hosting companies, security teams, and infrastructure leaders all have a role in understanding what comes next.
The conversations in Phoenix kept pointing back to the same idea, AI is touching every layer of the stack.
Security Is Going to Change Fast
Security came up repeatedly in our conversations because AI creates advantages for both defenders and attackers.
Security teams can use AI to detect suspicious behavior, automate analysis, and respond faster. Attackers can also use AI to find weaknesses, create convincing phishing attempts, generate malicious code, automate recon, and potentially scale attacks far beyond what was previously practical.
The speed of that change matters. Hosting providers may have less time to react when vulnerabilities become public, and traditional security workflows may not be fast enough in an environment where automated systems are acting continuously.
That makes the broader question of resilience even more important. Companies need to think not only about how to stop an attack, but also how they continue operating when something gets through.
What Happens When AI Goes Off the Rails?
This was another theme that kept coming up.
An AI agent does not have to be malicious to cause damage. It could receive a bad instruction, make an incorrect assumption, inherit too much access, or simply take an action that nobody expected.
The risk becomes more serious when those systems are connected directly to production infrastructure. A mistake that might once have affected a single website could potentially impact hundreds or thousands of environments if the automation is operating at scale.
That changes the way companies need to think about permissions, monitoring, recovery, and human oversight.
AI makes things faster, but speed cuts both ways.
That Is Exactly Why We Built Atlas Digital Summit
Atlas Digital Summit was created around the need for deeper conversations like these.
We did not want another large expo where everyone spends two days moving between booths, product demos, and 15-minute meetings. We wanted to create an environment where executives and technical leaders could actually sit down and work through the issues that are reshaping the industry.
What we experienced at WordCamp US reinforced that decision.
Hosting providers need to understand what security companies are seeing. Security leaders need to understand how AI agents are being deployed. Data center operators need to understand where compute demand is heading. Executives need to understand how all of these changes affect their customers, operations, margins, products, and long-term strategy.
Those conversations need more than a hallway between sessions.
AI Will Be Front and Center in Reykjavík
At Atlas Digital Summit, AI is not being treated as another trend to add to the agenda.
It is part of the core discussion because it affects almost everything else we are talking about, from infrastructure and security to customer support, automation, resilience, and business strategy.
Our roundtables will go deeper into agent safety, privileged access, credential theft, data exfiltration, AI-powered attacks, supply chain risks, infrastructure automation, and how hosting companies may need to redesign their organizations as AI becomes more capable.
We will also be talking about what happens when AI does not behave as expected, because that may become one of the most important operational questions in the next few years.
The Industry Needs a Room for This Conversation
WordCamp US gave us a glimpse of where the industry conversation is heading.
A discussion could start with WordPress, move into hosting, then into AI agents, infrastructure security, data centers, support automation, and the future of the hosting business. That happened repeatedly throughout the week.
Those conversations are already taking place, but they are often fragmented, a few minutes in a hallway, a dinner conversation, or a quick discussion between meetings.
Atlas Digital Summit is designed to give those conversations a room.
In Reykjavík, we are bringing together CEOs, founders, CTOs, security leaders, infrastructure operators, and AI experts in small moderated roundtables where the goal is not to pitch a product, but to understand what comes next.
Because the AI conversation is no longer coming.
It is already here, and WordCamp US made that very clear.