Vercel 通過 ISO/IEC 42001:2023 AI 管理系統認證
Vercel completes ISO/IEC 42001:2023 certification
企業把 AI 助理用在軟體開發時,也得說清楚誰負責、資料如何處理,以及系統上線後怎麼監看。
- Vercel 表示已通過 ISO/IEC 42001:2023 認證,範圍涵蓋底層基礎設施、AI Gateway、Vercel Agent、v0,以及相關人員和流程。這套標準要求組織評估 AI 系統可能帶來的好處、傷害與風險。
- Vercel 也表示會在重大變更時重新評估,並在發布前檢查系統。認證效期為三年,期間每年接受監督稽核。客戶可在方案允許的範圍內選擇資料處理方式,從零資料保留到不將資料用於訓練。
AI 伴隨問答
針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:
Vercel has completed its ISO/IEC 42001:2023 certification, meeting the international standard for AI management systems.
This certification gives our customers confidence to ship fast with agents on Vercel, demonstrating how Vercel designs, builds, operates, and governs our AI products.
Our AI management system covers Vercel’s underlying infrastructure and core AI products, including AI Gateway, Vercel Agent, v0, AI-based open source software and frameworks, as well as the people and processes behind them.
Vercel’s certificate is available here, and additional audit documentation is available through our Trust Center.
What is ISO 42001?
ISO/IEC 42001:2023 is the international standard for AI management systems and applies to any organization that develops, provides, or uses AI.
An AI management system is the set of policies, roles, processes, and controls through which an organization consistently governs AI activities. The standard includes requirements for leadership accountability, AI risk assessment, lifecycle management, data governance, transparency, and continual improvement.
The requirement that makes ISO 42001 distinct from security standards is the AI system impact assessment. Security frameworks cover how an organization protects information, but ISO 42001 covers how it builds and uses AI, including the assessment of how an AI system impacts the people it touches, and the benefits and harms that would result when that system materially changes.
To obtain the certification, we underwent a two-stage audit performed by an independent, accredited certification body, which evaluated whether our AI management system is designed correctly and operates effectively. The certificate is valid for three years, with annual surveillance audits in between.
“Enterprises want agents in production, but no one should have to choose between moving fast with AI and governing it well,” said Talha Tariq, CTO for Security. “We built our AI governance to run as continuously as we ship, and ISO 42001 is the independent check that it works.”
What this means for customers
AI moved into the critical path of how software gets built faster than the governance around it did. Teams are shipping agents, assistants, and model-backed features on Vercel every day, and the questions that follow are no longer hypothetical:
Who is accountable when a model behaves unexpectedly?
What data reached a model?
How is the system monitored once it’s live?
Who decided the system was ready?
We pursued ISO 42001 because we are committed to providing enterprise-grade AI products. This certification attests that we run deliberate, auditable AI systems.
How we govern AI at Vercel
There are three primary ways AI is governed at Vercel.
Impact assessments
Vercel conducts AI system impact assessments which cover foreseeable benefits, harms, failures, and misuse. We reassess these risks and benefits when the system materially changes.
Evaluations before release
We maintain evals for our AI systems covering performance, safety, and security, with human review for material changes.
Clear data commitments
Where it is not the default option based on pricing plan, we provide customers the ability to set their data handling preferences ranging from zero data retention (ZDR) to training opt-out. Our AI policy is available at v0.app/policy.
Ship fast, with confidence
We are committed to helping teams ship fast with confidence in the way AI is built and governed at Vercel. ISO 42001 is proof of that commitment, giving our customers:
Confidence in agentic infrastructure: Vercel’s AI systems have been audited, which means customers using our AI products start from a reviewed foundation instead of an unexamined one.
Governance can ship at the speed of the product: Agents now drive the development and deployment of applications. Governance and oversight that occurs quarterly can’t govern software that changes hourly, so we built a system that reviews AI as continuously as we ship it.
Simplified and accelerated assurance: Customers and partners can rely on this certification, alongside our other compliance attestations, to meet AI governance diligence requirements for faster third-party vendor onboarding.
Clear lines of responsibility: ISO 42001 requires that responsibilities be allocated across the AI supply chain. Our documentation outlines what Vercel is accountable for and what remains yours, the same shared-responsibility model we use for security.
Discipline that holds over time: Annual surveillance audits mean this isn’t a one-time claim. As our AI products change, the system that governs them stays under independent review.
AI is changing what people build on Vercel, and we intend for the governance around it to keep pace. We’ll continue to expand the scope of our certifications as our products grow.