AI Phishing Attacks Now 24% More Effective Than Humans: Here’s How to Fight Back

AI Phishing Attacks Now 24% More Effective Than Humans: Here’s How to Fight Back

From All Things Human Risk Management by Hoxhunt

July 22, 2025 · 28 min · Episode 4

About this episode

This episode discusses the effectiveness of AI in phishing attacks and how to combat them.

Episode #4 AI might be transforming cybersecurity… but not always in the way vendors want you to believe. In this episode, Eliot Baker sits down with Hoxhunt CTO and Co-founder Pyry Åvist to expose what’s actually happening at the front lines of the AI-powered phishing threat. No speculation. Just real research, real data, and real implications for your defense strategy. Together, they unpack how Hoxhunt’s research team built and tested agentic AI spear phishing agents and what happened when those AI-generated attacks went head-to-head with elite human red teamers. The result? A 24% higher failure rate for AI phishing emails, across 50,000+ real-world simulations. This is more than just a stat. It’s a signal: AI threats are getting smarter, faster than most training programs can adapt. This episode is both a behind-the-scenes look at the largest AI phishing benchmark ever run and a tactical guide for what to do next. Here’s what you’ll learn in this episode: How AI spear phishing attacks crossed a key threshold in spring 2025 and why that changes everything Why traditional training templates and static simulations are now a liability What “agentic” AI really means and how it’s…

People in this episode

Host: Eliot Baker

Guest: Pyry Åvist

Topics covered

  • AI
  • phishing
  • cybersecurity
  • technology

Keywords

  • AI phishing
  • cyber threats
  • defense strategy
  • agentic AI

Mentioned in this episode

Products: Hoxhunt

More episodes of All Things Human Risk Management

Explore listener stats, chart rankings, contacts and more on the All Things Human Risk Management podcast page.