AI Finds Ghostscript Upload Flaw Enabling Server-Side Code Execution
ccerrato147 · x · 2026-11-22
Paulos Yibelo used AI to help find a flaw in Ghostscript, a 38-year-old program that turns PDFs and print files into images behind many upload boxes; one crafted file runs code on the server. He demonstrated it on Debian's print system, Roundcube webmail, MediaWiki, WordPress plugins and a Google file renderer.
The post also carries a promo for pwn.ai, an autonomous penetration-testing product built by zero-day researchers: scope a target, let a swarm of agents crawl endpoints, parameters and auth flows, chain bugs and prove exploits, then return a pentest report with PoC code and remediation guidance. It lists research findings across Chrome, AWS, Linux, KVM, Firecracker, WordPress and more.
Related event: AI-assisted research uncovers Upload2Shell Ghostscript sandbox escape(5 posts)→
More from Safety
- Intruder swapped Prime Intellect's hosted DeepSeek for an abliterated copy for 53 hours — mawcopolow · 2026-11-22
- Dragos finds three hacked AI hosts whose operators had no idea they were breached — mawcopolow · 2026-11-22
- Who do you sue when an AI agent errs? The "sovereign agents" liability debate — QuixiAI · 2026-11-22
- Opinion: without provenance, 'ethical AI' is just a slide deck — AryHHAry · 2026-11-22
- OpenAI board member Christiano puts AI takeover risk at ~10% within a year — TheAIGRID · 2026-11-22
- OpenAI red team extracted protected reasoning from GPT-6 Astra 54% of the time; new safeguards cut it to 0.2% — AIFlow_ML · 2026-11-22