PhishGen Evaluation of LLM-Driven Phishing Website Generation

# PhishGen Evaluation of LLM-Driven Phishing Website Generation

https://gcdradar.vercel.app/incidents/phishgen-evaluation-of-llm-driven-phishing-website-generatio · AIR-2026-1475 · closed 2026-11-06

- Lane: Research
- Severity: 1 (Research / proof of concept)
- Fix status: Fix status unknown
- First reported: 2026-10-23
- AI role: AI-enabled attack

## Summary

Researchers evaluated the capabilities of large language models in automatically generating functional and visually deceptive phishing websites. The study benchmarks LLM-generated pages against human deception standards.

## What to do

No action recorded; read the sources below and the vendor's advisory.

## Latest updates

- 2026-10-23: PhishGen: Understanding LLM Capabilities in Automated Phishing Website Generation (https://arxiv.org/abs/2610.30226)

## Sources

- 2026-10-23 arXiv: PhishGen: Understanding LLM Capabilities in Automated Phishing Website Generation https://arxiv.org/abs/2610.30226

Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.