ExploitGym

Reported Context (3)

They found that one of the AI models participating in the attack included GPT-5.6 Sol with reduced cyber refusal guardrails for testing on ExploitGym, an AI benchmark of cyber capabilities. Hugging Face Breach Linked to AI Model Exploiting Vulnerabilities During Benchmark Testing
Once online, the models inferred that Hugging Face might contain models, datasets, or solutions associated with ExploitGym, the benchmark they had been instructed to solve. Hugging Face Incident Highlights Governance Risks of Autonomous AI Agents
findings automatically determined to be fixed. While GPT-5.5-Cyber outperformed GPT-5.5 on CyberGym, ExploitGym, and SEC-bench Pro. Security roundup: Klue breach exposes SaaS OAuth tokens, OpenAI expands patching initiative

People (7)

Vendors (26)

Products (23)

Tools (2)

Industries (4)

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.