Tiered Access to Frontier AI Models
Anthropic has implemented a stratified access model for its most advanced AI models, granting exclusive early access to Claude Mythos Preview only to the largest cybersecurity vendors. CrowdStrike, Palo Alto Networks, and Zscaler are the launch partners for Project Glasswing, giving them access to this entirely new class of AI model that Anthropic describes as too powerful for general release. These three companies, with combined valuations exceeding $290 billion and annual sales of $16.7 billion, can use Mythos Preview to discover complex vulnerabilities that earlier models missed entirely.
One rung below are vendors like SentinelOne and Trend Micro’s TrendAI unit, which can embed Anthropic’s most powerful generally available model, Claude Opus 4.7, in their tools but lack access to Mythos Preview. Opus 4.7 enables partners to identify logic flaws, non-linear attack paths, and supply-chain risks that follow unpredictable patterns. The tiered approach starkly contrasts with OpenAI’s more democratic strategy, where GPT-5.4-Cyber is accessible to a broader set of security partners through clear objective criteria and identity verification.
Market Impact and Partnership Dynamics
Anthropic’s dominance in the enterprise LLM API market has grown from 12% in 2023 to 32% by mid-2025, according to Menlo Ventures, largely at OpenAI’s expense. The AI lab is also pursuing services partnerships with Accenture, BCG, Deloitte, Infosys, and PwC to help organizations deploy Claude integrated security solutions. Meanwhile, CrowdStrike’s Project QuiltWorks, launched with Accenture, IBM, EY, Kroll, and OpenAI, targets the wave of vulnerabilities being discovered by frontier AI models in production code.
This partnership strategy creates a clear hierarchy where the most valuable cybersecurity companies earn privileged access to the most advanced defensive AI capabilities. Zscaler will integrate Claude Mythos Preview into its secure software development lifecycle to accelerate vulnerability discovery in its Zero Trust Exchange platform. Major cyber vendors are also building their own AI security services teams, with Palo Alto Networks collaborating with Accenture, Deloitte, IBM, NTT Data, and PwC to drive enterprise AI resilience.
Source: Healthcareinfosecurity
