<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Api-Risk on BRYSGO</title><link>https://www.brysgo.com/tags/api-risk/</link><description>Recent content in Api-Risk on BRYSGO</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 29 Jul 2026 16:06:41 +0000</lastBuildDate><atom:link href="https://www.brysgo.com/tags/api-risk/index.xml" rel="self" type="application/rss+xml"/><item><title>The API You Depend On Can Be Switched Off by Export Controls</title><link>https://www.brysgo.com/post/2026-07-29-the-api-you-depend-on-can-be-switched-off-by-export-controls/</link><pubDate>Wed, 29 Jul 2026 16:06:41 +0000</pubDate><guid>https://www.brysgo.com/post/2026-07-29-the-api-you-depend-on-can-be-switched-off-by-export-controls/</guid><description>&lt;p&gt;Anthropic didn&amp;rsquo;t choose to take its best models offline globally. On June 12, 2026, the U.S. Commerce Department&amp;rsquo;s Bureau of Industry and Security ordered it to — because a jailbreak had exposed a vulnerability that could help identify cyber targets, and because Anthropic couldn&amp;rsquo;t verify user nationality fast enough to comply with the resulting directive. The models went dark for everyone. Not just adversaries. Everyone.&lt;/p&gt;
&lt;p&gt;If you build on top of frontier AI APIs, that should stop you cold.&lt;/p&gt;</description></item></channel></rss>