Veracode Report Highlights First Significant Improvement in AI-Generated Code Security
Veracode released a report detailing the security of AI-generated code, indicating the first significant improvement in this area. The report, widely covered by multiple tech outlets, states that the average pass rate for security checks on AI-generated code has reached 62%. Leading models, specifically Claude Opus 5.5 and GPT-6-Sol, achieved a 73% pass rate. While acknowledging these gains, the report and subsequent coverage emphasize that security gaps persist, and organizations should not solely rely on AI for application security. The central claims of improvement and specific pass rates originate from Veracode's own benchmark findings.
7 articles from 7 outlets covered this story. Their coverage differs on 3 points. The underlying claim is sourced from a benchmark.
What do all outlets agree on?
7 outlets covered “Veracode Report Highlights First Significant Improvement in AI-Generated Code Security”. All of them report the following:
- AI-generated code security has improved
- Veracode published a report on AI code security
- Leading AI models include Claude and GPT-6 variants
- Security vulnerabilities or gaps still exist in AI-generated code
Did outlets disagree about this?
Yes. Coverage of “Veracode Report Highlights First Significant Improvement in AI-Generated Code Security” differs on 3 points. Each account below is how a different outlet described the same event:
The specific timeframe for the 'first significant improvement'
The precise naming of the leading AI models
The extent to which organizations can rely on AI for application security
Which outlets covered this?
All 7 articles found on this story, grouped by the stance of the piece. Every link goes to the original publisher.