OpenAI has rolled out GPT-5.5-Cyber, a specialized model for cybersecurity teams, via a controlled-access program. The launch follows Anthropic's Mythos debut by roughly one month, intensifying competition for AI-native security tooling.
Cyber-specialist models reflect maturation of domain-specific AI. Both vendors are betting on secure, auditable access patterns to earn trust with InfoSec buyers who operate under compliance regimes (SOC 2, FedRAMP). Vendor lock-in dynamics here favor incumbents with existing enterprise relationships.