Industry

Harmonic CEO:多数安全漏洞来自遗留代码而非 AI 生成代码

Why we picked it

Harmonic 的 Vlad Tenev 用 CVE 数据库数据说明漏洞大头其实是遗留代码,不是 AI 写的。他还讲了形式化验证怎么在代码合并前把漏洞数学化排除掉。做安全审计或用 AI 编程工具的团队都该看看这个视角。

Harmonic CEO Vlad Tenev(@tachim)接受 The Information 采访时表示,AI 生成代码正引发新的安全担忧,但他在公共 CVE 数据库中追踪到的漏洞大多来自遗留代码,而非 AI 新生成的代码。这一判断与当前普遍将软件供应链风险归因于 AI 编程工具的叙事形成反差。

Body

Tenev 解释称,Harmonic 正在推进形式化验证(formal verification)方法,用数学方式证明代码不存在特定类别的漏洞,帮助开发者在合并代码前定位并消除隐患。形式化验证在硬件设计领域已成熟,但应用到日常软件开发的工具链仍处于早期阶段。

对开发团队而言,这一观点提示安全审计的资源应优先投向存量遗留系统,而非简单禁止 AI 编码工具。随着 AI 生成代码占比上升,形式化验证类工具可能成为 CI/CD 流程中的新环节。

Original · The Information

AI-generated code is raising new security concerns. But is it responsible for the vulnerabilities we're seeing?

@HarmonicMath CEO @tachim says many of the vulnerabilities he's tracking come from older software:

“What we see is that in the public CVE database, most of the vulnerabilities I'm tracking come from legacy code, not from new code.”

He explains how formal verification could help developers identify and eliminate certain vulnerabilities.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Industry" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Harmonic, formal verification, AI safety, CVE.

Source attribution: curated from The Information (@theinformation).