GPT-5.5-Cyber Limited Preview Launched, Specializing in Vulnerability Identification and Malware Analysis
15 hour ago / Read about 0 minute
Author:小编   

OpenAI announced on Thursday that its latest AI model, GPT-5.5's cybersecurity-focused variant, GPT-5.5-Cyber, has begun offering a limited preview to vetted security teams. This comes a month after its competitor, Anthropic, released the Claude Mythos preview, a model that garnered significant attention from investors and government officials for its ability to autonomously discover software vulnerabilities. OpenAI stated that GPT-5.5-Cyber is designed to lift restrictions on security task execution, helping teams more efficiently carry out tasks such as vulnerability identification, patch verification, and malware analysis.