Effective immediately, the Zhongguancun Independent Large Model Industry Alliance officially launches the “Cybersecurity Professionals Open Program”, offering trusted security partners invited, controlled access to the capabilities of self-reliant large models in cybersecurity scenarios. The program follows the principles of lawful authorization, defense-first and auditable processes; it is not open to the public, but provides a controlled access mechanism to trusted defenders.
The Alliance hopes to help trusted security partners — under the premises of lawful authorization and defense-first — improve efficiency in vulnerability discovery, code auditing, patch validation, security testing, code repair and secure development, putting advanced large-model capabilities to work for cyber-defense. Alliance member model vendors provide model capabilities and technical support, while Alliance partners are responsible for compliant use scenarios, target authorization, result review and vulnerability handling.
Who it is open to
- Relevant cybersecurity departments and supporting units
- Security research institutions, key laboratories and university security teams
- Cybersecurity enterprises, software-security service providers and infrastructure-security teams
- Security researchers and open-source infrastructure maintainers approved by the Alliance
Partner benefits
- Access to self-reliant large-model capabilities and enterprise-grade technical support, including quota and concurrency guarantees, integration and scenario adaptation
- Closed-door exchange, joint evaluation and scenario co-creation opportunities within the Alliance
- Joint release and industry dissemination support for outstanding results
Admission principles
- Invitation and verification: submit a defensive use-scenario description; access is granted after Alliance review
- Lawful authorization, defense-first: use is limited to authorized scenarios; abnormal calls and suspected violations are governed through auditing and access control
- Human review: model output is expert assistance rather than a final conclusion, and must undergo human review and engineering validation before use in production handling or external disclosure; vulnerabilities follow a responsible-disclosure process
Interested enterprises, institutions or individuals can submit applications and defensive use-scenario descriptions via the form below. Submissions enter a manual review process, usually completed within 3 business days, with results notified by email.