AI Security Risks Are Changing How MSPs Protect the Inbox

When Channel Insider covered the AI security conversation at GTIA ChannelCon 2026, one of the voices they turned to on email was Mailprotector Founder and CEO David Setzer. The reporting, by managing editor Victoria Durgin, framed AI as a two-sided problem for MSPs. Attackers are using it to make old attacks more convincing, and customers are adopting AI tools and quietly exposing confidential data.

Setzer characterized AI as “human intelligence at scale.” Point that scale at an email phishing campaign and the result is a message with clean language, correct branding, and none of the mistakes users were ever trained to catch. As he put it to Channel Insider:

“AI creates the perfect phishing message, with perfect grammar, perfect colloquialisms and perfect fake logos. If you look at phishing messages now, you cannot distinguish them from a message from a real company.”

Setzer’s perspective is featured alongside the Global Technology Industry Association (GTIA)‘s MJ Shoer, MSP Founder and Wake Technical Community College professor Sandellyo Kauba, and Coro’s Demetrios Georgiou. Explore their unique points of view and the full ChannelCon 2026 report by reading AI Security Risks Push MSPs to Rethink Customer Protection by Victoria Durgin at Channel Insider.

What This Means for MSP Email Security Practices

The Channel Insider article highlights a larger challenge for MSPs: as AI makes traditional phishing indicators less reliable, email security strategies have to evolve with it. Below, we answer a few of the questions we’re hearing from MSPs about what that means for email protection.

Can users still spot AI-generated phishing emails?

Not reliably, based on traditional tells. As David Setzer told Channel Insider, AI now produces messages with perfect grammar, colloquialisms, and fake logos that can’t be distinguished from real company email. The indicators most mainstream awareness trainings were built around were byproducts of human effort that ultimately no longer show up in AI-driven attacks.

If detection no longer works, what stops AI phishing?

Changing the default assumption. Zero trust email security does not treat any email as automatically trustworthy and makes every message earn its way to the mailbox, rather than granting delivery and trying to catch what’s bad afterward.

Does AI-powered email security really scan every message with AI?

Not at scale. Mailprotector’s position is that the compute cost makes full AI scanning of every message economically impractical, and that many AI-powered claims describe machine learning that has been in email filters for years.

How does Mailprotector Shield use AI?

Shield uses AI where it makes email safer or simpler to manage, such as summarizing messages in Bundler so you can quickly decide what deserves a closer read. We call this intentional AI: choosing the right technology for each task. Shield’s Zero Trust security model controls who reaches your inbox; AI supports specific features within that foundation. Explore our patents, including those related to AI.

How can an MSP check a client’s current email security posture?

Radar runs live round-trip email tests and returns a risk score at no cost. Learn more at Radar.