<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[数据分级表写出来了，AI 系统怎么真的执行]]></title><description><![CDATA[<p dir="auto">公司有数据分级表，但 AI 系统接入时没人知道怎么落地。表上写“内部敏感”，代码里怎么体现？</p>
]]></description><link>https://localaihub.com/topic/151/数据分级表写出来了-ai-系统怎么真的执行</link><generator>RSS for Node</generator><lastBuildDate>Wed, 03 Jun 2026 19:23:07 GMT</lastBuildDate><atom:link href="https://localaihub.com/topic/151.rss" rel="self" type="application/rss+xml"/><pubDate>Sun, 10 May 2026 15:54:00 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 16:04:00 GMT]]></title><description><![CDATA[<p dir="auto">数据分级不是 Excel，是每条链路上的字段和判断。</p>
]]></description><link>https://localaihub.com/post/1549</link><guid isPermaLink="true">https://localaihub.com/post/1549</guid><dc:creator><![CDATA[melo]]></dc:creator><pubDate>Mon, 11 May 2026 16:04:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 14:03:00 GMT]]></title><description><![CDATA[<p dir="auto">先做保留周期和脱敏。别等安全来找。</p>
]]></description><link>https://localaihub.com/post/1548</link><guid isPermaLink="true">https://localaihub.com/post/1548</guid><dc:creator><![CDATA[小吴]]></dc:creator><pubDate>Mon, 11 May 2026 14:03:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 11:36:00 GMT]]></title><description><![CDATA[<p dir="auto">我们日志现在基本全量。</p>
]]></description><link>https://localaihub.com/post/1547</link><guid isPermaLink="true">https://localaihub.com/post/1547</guid><dc:creator><![CDATA[小满]]></dc:creator><pubDate>Mon, 11 May 2026 11:36:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 09:05:00 GMT]]></title><description><![CDATA[<p dir="auto">最容易漏的是回放日志。你以为只是调试，里面可能有原文片段和用户输入。</p>
]]></description><link>https://localaihub.com/post/1546</link><guid isPermaLink="true">https://localaihub.com/post/1546</guid><dc:creator><![CDATA[阿白]]></dc:creator><pubDate>Mon, 11 May 2026 09:05:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 06:22:00 GMT]]></title><description><![CDATA[<p dir="auto">还有输出。模型可能把敏感信息组合出来，输出前也要检查。</p>
]]></description><link>https://localaihub.com/post/1545</link><guid isPermaLink="true">https://localaihub.com/post/1545</guid><dc:creator><![CDATA[陈一]]></dc:creator><pubDate>Mon, 11 May 2026 06:22:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 05:35:00 GMT]]></title><description><![CDATA[<p dir="auto">两个都可以。身份证、手机号、邮箱这种规则就够；业务敏感词可能要模型辅助，但不能全靠模型。</p>
]]></description><link>https://localaihub.com/post/1544</link><guid isPermaLink="true">https://localaihub.com/post/1544</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Mon, 11 May 2026 05:35:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 04:27:00 GMT]]></title><description><![CDATA[<p dir="auto">检测靠规则还是模型？</p>
]]></description><link>https://localaihub.com/post/1543</link><guid isPermaLink="true">https://localaihub.com/post/1543</guid><dc:creator><![CDATA[小蓝]]></dc:creator><pubDate>Mon, 11 May 2026 04:27:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 01:54:00 GMT]]></title><description><![CDATA[<p dir="auto">要。用户可能把客户手机号、合同金额直接问进去。输入也要做检测和日志策略。</p>
]]></description><link>https://localaihub.com/post/1542</link><guid isPermaLink="true">https://localaihub.com/post/1542</guid><dc:creator><![CDATA[Grace]]></dc:creator><pubDate>Mon, 11 May 2026 01:54:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Mon, 11 May 2026 00:33:00 GMT]]></title><description><![CDATA[<p dir="auto">用户问题本身也要分级吗？</p>
]]></description><link>https://localaihub.com/post/1541</link><guid isPermaLink="true">https://localaihub.com/post/1541</guid><dc:creator><![CDATA[普通网友A]]></dc:creator><pubDate>Mon, 11 May 2026 00:33:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Sun, 10 May 2026 23:15:00 GMT]]></title><description><![CDATA[<p dir="auto">不然 RAG 检索时没法过滤。你不能等答案生成后再想它用了什么资料。</p>
]]></description><link>https://localaihub.com/post/1540</link><guid isPermaLink="true">https://localaihub.com/post/1540</guid><dc:creator><![CDATA[阿航]]></dc:creator><pubDate>Sun, 10 May 2026 23:15:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Sun, 10 May 2026 20:53:00 GMT]]></title><description><![CDATA[<p dir="auto">那第一步是给数据源打标签。文档、chunk、用户、会话、工具结果都要能带 classification。</p>
]]></description><link>https://localaihub.com/post/1539</link><guid isPermaLink="true">https://localaihub.com/post/1539</guid><dc:creator><![CDATA[林小北]]></dc:creator><pubDate>Sun, 10 May 2026 20:53:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Sun, 10 May 2026 18:49:00 GMT]]></title><description><![CDATA[<p dir="auto">现在只有文档，没有系统字段。</p>
]]></description><link>https://localaihub.com/post/1538</link><guid isPermaLink="true">https://localaihub.com/post/1538</guid><dc:creator><![CDATA[小满]]></dc:creator><pubDate>Sun, 10 May 2026 18:49:00 GMT</pubDate></item><item><title><![CDATA[Reply to 数据分级表写出来了，AI 系统怎么真的执行 on Sun, 10 May 2026 16:35:00 GMT]]></title><description><![CDATA[<p dir="auto">先把分级变成策略：能否进入云模型、能否进入本地模型、能否进日志、能保留多久、谁能看。</p>
]]></description><link>https://localaihub.com/post/1537</link><guid isPermaLink="true">https://localaihub.com/post/1537</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Sun, 10 May 2026 16:35:00 GMT</pubDate></item></channel></rss>