Skip to content

DEV Community

AI Security Benchmark Series Series' Articles

Back to Ofri Peretz's Series

Cover image for I Let Claude Write 80 Functions. 65-75% Had Security Vulnerabilities.

Feb 6

I Let Claude Write 80 Functions. 65-75% Had Security Vulnerabilities.

#ai #security #javascript #devsecops

32 min read

Cover image for I Asked Claude to Fix Its Own Security Bugs. 1 in 3 Fixes Added a NEW Vulnerability.

Feb 8

I Asked Claude to Fix Its Own Security Bugs. 1 in 3 Fixes Added a NEW Vulnerability.

#ai #security #javascript #devsecops

26 min read

Cover image for We Ranked 5 AI Models by Security. The Leaderboard Is Wrong.

Feb 11

We Ranked 5 AI Models by Security. The Leaderboard Is Wrong.

#ai #security #javascript #devsecops

15 min read

Cover image for Aggregate Benchmarks Lie. Here's What 700 AI Functions Look Like by Security Domain.

May 17

Aggregate Benchmarks Lie. Here's What 700 AI Functions Look Like by Security Domain.

#ai #security #eslint #devsecops

27 min read

Cover image for Claude Wrote a NestJS Service. TypeScript Was Happy. ESLint Found 6 Security Holes.

May 29

Claude Wrote a NestJS Service. TypeScript Was Happy. ESLint Found 6 Security Holes.

#ai #security #nestjs #eslint

20 min read

Cover image for Same NestJS Prompt. Claude Got 6 Security Errors. Gemini Got 2. Here's What Both Got Wrong.

May 30

Same NestJS Prompt. Claude Got 6 Security Errors. Gemini Got 2. Here's What Both Got Wrong.

#ai #security #googleai #geminichallenge

13 min read

Cover image for Claude vs Gemini Across 4 Security Domains: A Dead Heat — and the Hardening 63% of AI Code Skips

May 31

Claude vs Gemini Across 4 Security Domains: A Dead Heat — and the Hardening 63% of AI Code Skips

#ai #security #googleai #eslint

9 min read