Building AI security in the open

GitLab CEO Bill Staples argues that as AI makes software production cheaper and more abundant, trust, verification, governance, context, and security become increasingly important, particularly because organizations must trust both AI agents and the code they generate. GitLab is joining the Open Secure AI Alliance to help develop open security technologies and practices, emphasizing that AI security benefits from community inspection, testing, adaptation, and improvement. The company highlights its publicly available AI Gateway source code and agent workflow configurations as examples of transparency, while stressing that trustworthy AI depends on the full system of workflows, controls, context, and evaluation rather than only the underlying model. GitLab also advocates model and cloud flexibility, allowing enterprises to choose proprietary or open-weight models and deployment approaches, and describes its agents as privileged actors with explicit identities, limited permissions, sandboxed execution, auditable actions, and governed tools.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.