Tool
ExploitGym
- First Reported
- Jul 21, 2026
- Latest Reported
- Nov 6, 2026
Reported Context (28)
- It also scores 48.5% on ExploitGym, a cybersecurity benchmark that evaluates the capabilities of AI agents to autonomously turn known software vulnerabilities into functioning, real-world exploits. In comparison, OpenAI Launches GPT-6 Cyber, Citing 81 Firefox Flaws and 948 Linux Privilege-Escalation Exploits
- On ExploitGym, which checks whether an AI can exploit a given flaw in a remote target, the new model achieved 48.5%, up from 32.1% for GPT-5.6 Cyber. OpenAI Unveils GPT-6 Cyber for Authorized Vulnerability Research
- Illustration of GPT-6 training flow for staying in-scope during evaluations on ExploitGym. OpenAI unveils GPT-6 Cyber, touting vulnerability research and safety gains
- GPT-6 Cyber scored 48.5% on ExploitGym, which measures if models can exploit certain vulnerabilities, higher than the 32.1% posted by GPT-5.6 Cyber. In alignment evaluations, the new model avoided answering 99.93% of OpenAI Unveils GPT-6 Cyber, Citing Firefox Vulnerabilities and Exploit Gains
- The models were participating in an internal cyber security evaluation using ExploitGym, a benchmark designed to assess offensive cyber capabilities. NCC Group Reports August Ransomware Rise and Risks From AI Evaluation Failures
CVE (10)
People (31)
Threat Actors (6)
MITRE ATT&CK (25)
Vendors (22)
Products (65)
Tools (9)
Industries (11)
Countries (18)
Note: Related entities, including threat actors, malware, CVEs, MITRE ATT&CK techniques, vendors, products, tools, countries, and industries, are shown when they appear in the same reporting. Their presence does not necessarily mean they were targeted, compromised, vulnerable, responsible for the activity, or directly involved in the incident.
