<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>SAPIEN Framework Blog</title><description>Updates, research notes, and announcements from the SAPIEN Behavioral Safety Framework.</description><link>https://sapienframework.org/</link><item><title>Nobody Measures What Happens After Turn Five</title><link>https://sapienframework.org/blog/nobody-measures-what-happens-after-turn-five/</link><guid isPermaLink="true">https://sapienframework.org/blog/nobody-measures-what-happens-after-turn-five/</guid><description>We surveyed the 2026 research landscape on AI behavioral drift — sycophancy benchmarks, multi-turn evals, judge-panel research — and mapped the five-part gap that, to our knowledge, no published benchmark covers.</description><pubDate>Mon, 06 Jul 2026 00:00:00 GMT</pubDate></item><item><title>The SAPIEN Certification Program Is Open</title><link>https://sapienframework.org/blog/sapien-certification-launch/</link><guid isPermaLink="true">https://sapienframework.org/blog/sapien-certification-launch/</guid><description>The SAPIEN Certification Program is now open: training and credentialing for individuals in AI behavioral safety assessment, built on the published SAPIEN methodology.</description><pubDate>Sat, 04 Jul 2026 00:00:00 GMT</pubDate></item><item><title>How Claude Sonnet 5 Scores on SAPIEN</title><link>https://sapienframework.org/blog/how-claude-sonnet-5-scores-on-sapien/</link><guid isPermaLink="true">https://sapienframework.org/blog/how-claude-sonnet-5-scores-on-sapien/</guid><description>Anthropic&apos;s newest agentic model posts a Health Score of 87 and ranks #2 on our council-scored board. But a strong average hides a High-risk tail. Here&apos;s the full breakdown.</description><pubDate>Wed, 01 Jul 2026 00:00:00 GMT</pubDate></item><item><title>The Chinese Model Problem</title><link>https://sapienframework.org/blog/the-chinese-model-problem/</link><guid isPermaLink="true">https://sapienframework.org/blog/the-chinese-model-problem/</guid><description>Three of the six AI models we tested are built by companies legally required to cooperate with Chinese state intelligence. They also happen to be the least reliable.</description><pubDate>Fri, 17 Apr 2026 00:00:00 GMT</pubDate></item><item><title>We Tested 6 AI Models. Most of Them Caved.</title><link>https://sapienframework.org/blog/we-tested-6-models/</link><guid isPermaLink="true">https://sapienframework.org/blog/we-tested-6-models/</guid><description>792 scenario runs across 11 risk domains. Every model was told to hold a safety boundary. Here&apos;s which ones did and which ones didn&apos;t.</description><pubDate>Fri, 17 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Your AI Is Agreeing With You. That&apos;s the Problem.</title><link>https://sapienframework.org/blog/your-ai-is-agreeing-with-you-sapien/</link><guid isPermaLink="true">https://sapienframework.org/blog/your-ai-is-agreeing-with-you-sapien/</guid><description>A Stanford study just confirmed what practitioners already knew: AI chatbots are built to tell you what you want to hear. Here&apos;s what that means, why it&apos;s measurable, and how to see it for yourself.</description><pubDate>Wed, 15 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Why AI Sycophancy Matters for Managed Service Providers</title><link>https://sapienframework.org/blog/why-sycophancy-matters-for-msps/</link><guid isPermaLink="true">https://sapienframework.org/blog/why-sycophancy-matters-for-msps/</guid><description>MSPs deploy AI tools to businesses that depend on them. When those tools drift under pressure, the consequences land on real people.</description><pubDate>Fri, 20 Mar 2026 00:00:00 GMT</pubDate></item><item><title>Introducing SAPIEN v1.1</title><link>https://sapienframework.org/blog/introducing-sapien-v1-1/</link><guid isPermaLink="true">https://sapienframework.org/blog/introducing-sapien-v1-1/</guid><description>Major expansion of the SAPIEN Behavioral Safety Framework — 14 pressure techniques, rapport as a distinct drift mode, and formal conformance requirements.</description><pubDate>Sun, 15 Mar 2026 00:00:00 GMT</pubDate></item></channel></rss>