Grounded or Gamed? We Audited Our Own Cyber Benchmark

Semgrep audited its own LLM benchmark for IDOR vulnerability detection after a viral model score raised doubts about whether models truly reason or just pattern match. It built four groundedness metrics—counterfactual (patch the bug, the finding should vanish), metamorphic (rename identifiers, the finding should persist), selectivity (don't flag safe look-alikes), and stability across repeated scans—and tested Kimi K2.7, GLM-5.2, MiniMax M3, Opus 4.8, GPT-5.5, and its own Semgrep Multimodal on its hardest repo. Verdict: nobody is gaming the benchmark—counterfactual and selectivity scores are near 1, so findings are causal rather than surface matches. The real problem is recall: of 27 real IDORs, 19 (70%) were missed by every model, and pooling all five models bought only one extra vulnerability. Misses split into depth (authorization logic spread across the request path) and familiarity (models catch the textbook file-download IDOR but miss the identical bug on other endpoints). A preview shows GLM stays grounded on SSRF but coverage collapses from 22% to 5.6%. The post is candid about limits—one repo, varying harnesses, no precision re-scoring—and concludes grounding is table stakes while recall is the frontier. 

https://semgrep.dev/blog/2026/grounded-or-gamed-we-audited-our-own-cyber-benchmark/

Comments

Post a Comment

Popular posts from this blog

OWASP ASVS 5.0 Released - Key Updates and What You Need to Know

The OWASP Foundation has released version 5.0 of the Application Security Verification Standard (ASVS), a major update to their security framework for web applications. This new version features restructured security requirements for better clarity, expanded guidance for cloud and API security, improved DevSecOps integration for CI/CD pipelines, updated threat modeling support, and enhanced compliance mapping with standards like NIST and PCI DSS. ASVS serves as a critical benchmark for developers building secure applications, penetration testers conducting security assessments, and auditors performing compliance reviews.  The standard is available for download from the OWASP ASVS project page, with organizations encouraged to integrate it into their software development lifecycles through code reviews and security testing tools. As a vendor-neutral, community-driven project, OWASP continues to welcome contributions to further develop the standard. This release represents an importa...
Read more

Critical OpenSSH Flaws Enable MITM and DoS Attacks

On February 18, 2025, HackRead reported on two critical vulnerabilities in OpenSSH, identified by the Qualys Threat Research Unit (TRU). The first vulnerability, CVE-2025-26465, affects the OpenSSH client and permits machine-in-the-middle attacks, potentially allowing attackers to impersonate legitimate servers and compromise SSH session integrity. This flaw exists regardless of the 'VerifyHostKeyDNS' setting and has been present since OpenSSH version 6.8p1. The second vulnerability, CVE-2025-26466, impacts both the client and server, enabling pre-authentication denial-of-service attacks that consume excessive system resources, leading to potential outages. Introduced in version 9.5p1, this issue persists up to version 9.9p1. Users are strongly advised to upgrade to OpenSSH version 9.9p2 to mitigate these vulnerabilities.  https://hackread.com/critical-openssh-flaws-expose-users-mitm-dos-attacks/
Read more

MITRE ATT&CK v19 Redefines How Defenders Model Modern Threats

MITRE ATT&CK v19 introduces one of the framework’s most significant structural changes in years, splitting the former Defense Evasion tactic into two clearer categories: Stealth and Defense Impairment. The release also expands coverage for AI-enabled adversary behavior, social engineering, and mobile detection strategies, while adding greater precision to ICS through new sub-techniques. Beyond taxonomy updates, v19 signals a broader evolution of threat modeling—toward more actionable, behavior-driven intelligence that reflects how attackers increasingly blend automation, deception, and cross-domain operations.  https://medium.com/mitre-attack/attack-v19-ff329cb65d66
Read more