Semgrep says GLM 5.2 beat Claude in a narrow security benchmark
The result strengthens Isaac Evans' long-running bet that code security needs better workflow design, not just bigger models.
By Ryan Merket · Published
Primary source: Semgrep
Why it matters
Semgrep's benchmark points to a shift in AI security: the defensible layer may be the workflow around the model, not the model itself.
The result strengthens Isaac Evans' long-running bet that code security needs better workflow design, not just bigger models.
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.