Tuesday, August 11, 2026

The worst AI model in our 2026 dataset fails half its security tests.

 
 
 
Report
 
2026 GenAI Code Security Report
 
11 new models tested across 80 tasks. Here's what the data means for your AI adoption decisions.
 
 
 

Coding models purpose-built for speed average 51% on security tasks. General-purpose models average 52%. There's no safety advantage to the tools marketed specifically at developers.

GPT-5.5 leads the Summer 2026 dataset at 68%. The worst performer fails on 1 in 2 security tasks. At today’s code volume production, that 18-point spread means model choice matters - a lot.

Four years of data. Over 100 models. No sign of a security breakthrough. If your team is scaling AI-assisted development, our 2026 GenAI Code Security Report tells you which tools to trust and which ones to watch.
 
 
 
 
 
 
 
 
 
at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)

Vantage generates $23m of ILS fee income in Q2

It's the first time we've seen a figure for fee income generated by the Vantage Partnership Capital (AdVantage) strategies ‌ ‌ ‌ ‌...