Obiguard launches Governance AI for guardrails, evaluations and audit trails across AI agents
Policy enforcement, Project Moonshot red-teaming, nightly bias scoring and audit trails for every AI agent, with controls mapped to five AI frameworks.
新闻稿以英文发布。
KUALA LUMPUR — Obiguard today launched Governance AI, a product for governing every AI agent an organisation runs.
Governance AI enforces guardrails and policy on prompts and completions, runs Project Moonshot red-teaming against deployed models, and scores bias nightly so drift is caught between reviews. Every decision is written to an append-only audit trail that auditors and regulators can inspect.
Controls map to five AI frameworks: NIST AI RMF, ISO/IEC 42001, OWASP LLM Top 10, Malaysia’s AISCF and Singapore’s MGF for GenAI, so teams can show evidence against the standards they are measured on.
Obiguard is a deep-tech startup founded in Malaysia in 2024 and backed by Antler. Governance AI is available now. Details are at obiguard.ai/governance.
Obiguard 是一家于 2024 年在马来西亚创立的深度科技初创公司,获得 Antler 支持。我们构建 AI 安全与治理基础设施,帮助组织审查 AI 使用情况、执行政策,并留存监管机构所期望的审计证据。了解更多:obiguard.ai/about。
媒体联系:[email protected]