<?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[Receptiviti Labs]]></title><description><![CDATA[AI systems have no validated view of what they're doing to users. Receptiviti Labs is building it.]]></description><link>https://labs.receptiviti.com/library</link><generator>RSS for Node</generator><lastBuildDate>Sat, 25 Jul 2026 13:06:05 GMT</lastBuildDate><atom:link href="https://labs.receptiviti.com/blog-feed.xml" rel="self" type="application/rss+xml"/><item><title><![CDATA[Why Model Tuning Alone Won't Fix Mental Health AI Safety]]></title><description><![CDATA[Tuning models toward expert judgment won't make mental-health AI safe: experts rate the same responses differently, and an un-inspectable model still makes the decisions.]]></description><link>https://labs.receptiviti.com/post/why-model-tuning-wont-fix-mental-health-ai-safety</link><guid isPermaLink="false">6a5c4e76e447e3bde588c01a</guid><category><![CDATA[Labs]]></category><pubDate>Mon, 20 Jul 2026 12:00:18 GMT</pubDate><enclosure url="https://static.wixstatic.com/media/e0675d_a8a38166f6da4728974ac48a895b60da~mv2.png/v1/fit/w_1000,h_630,al_c,q_80/file.png" length="0" type="image/png"/><dc:creator>Receptiviti Labs</dc:creator></item><item><title><![CDATA[AI Can Cause Harm: The Case for Psycholinguistic AI Safety]]></title><description><![CDATA[Ziv Ben-Zion's Nature piece calls for AI systems to flag distress. The catch: distress doesn't show up in a single message. It shows up as evidenced psychological shifts in how a person's language changes across a conversation, and catching that takes psycholinguistic measurement, not a keyword scan.]]></description><link>https://labs.receptiviti.com/post/psycholinguistic-ai-safety</link><guid isPermaLink="false">6a5a8ae15e7d848635df7f3d</guid><category><![CDATA[Labs]]></category><pubDate>Fri, 17 Jul 2026 21:47:54 GMT</pubDate><enclosure url="https://static.wixstatic.com/media/e0675d_d0e1f953f58440a4b16e9d309d138e1a~mv2.png/v1/fit/w_1000,h_1000,al_c,q_80/file.png" length="0" type="image/png"/><dc:creator>Receptiviti Labs</dc:creator></item><item><title><![CDATA[AI Can't Tell If It's Helping Your Judgment or Replacing It]]></title><description><![CDATA[A model can be helpful in every single response and still leave someone worse off by the tenth session — a little less able to think things through on their own, a little more ready to defer. No model can see that pattern, because it never shows up inside any one conversation. Here's the case for measuring it directly — and why building AI that extends human judgment, not just automates it, needs an instrument to prove it's working.
]]></description><link>https://labs.receptiviti.com/post/ai-over-reliance-judgment</link><guid isPermaLink="false">6a514a1e5c421b82021ab993</guid><category><![CDATA[Labs]]></category><pubDate>Fri, 10 Jul 2026 19:55:32 GMT</pubDate><enclosure url="https://static.wixstatic.com/media/e0675d_8add34a149414bfcadcbfc6bb9e0d614~mv2.png/v1/fit/w_1000,h_1000,al_c,q_80/file.png" length="0" type="image/png"/><dc:creator>Receptiviti Labs</dc:creator></item><item><title><![CDATA[What AI Infers About Users Is Part of the Interpretability Gap]]></title><description><![CDATA[A recent Science editorial argues we're losing the ability to understand AI. One piece of the problem is more tractable than it looks. The key is measurement, and inverting what gets inspected: not the model's output, but the user's language.]]></description><link>https://labs.receptiviti.com/post/ai-interpretability-measuring-the-user</link><guid isPermaLink="false">6a4b3053fa58f0a255d68531</guid><category><![CDATA[Labs]]></category><pubDate>Wed, 08 Jul 2026 04:44:22 GMT</pubDate><enclosure url="https://static.wixstatic.com/media/e0675d_9ba6daac9bb2474e9c218619e74519a8~mv2.png/v1/fit/w_1000,h_627,al_c,q_80/file.png" length="0" type="image/png"/><dc:creator>Receptiviti Labs</dc:creator></item><item><title><![CDATA[Why We Built Receptiviti Labs]]></title><description><![CDATA[The human-AI interaction is psychological in nature, and language carries that signal. AI systems already infer your cognitive and emotional state from language and respond to it — but that inference lives inside the model, with no way to verify it, and no external check on whether the system is reading you right. Receptiviti Labs measures that state directly, independent of the model — AI user state measurement grounded in psycholinguistic science.]]></description><link>https://labs.receptiviti.com/post/why-we-built-receptiviti-labs-ai-user-state-measurement</link><guid isPermaLink="false">6a4bc4cd2ca04bfb9dbad041</guid><category><![CDATA[Labs]]></category><pubDate>Sun, 05 Jul 2026 15:58:52 GMT</pubDate><enclosure url="https://static.wixstatic.com/media/e0675d_1340beffd1634b63bac69b6f6c27d03c~mv2.png/v1/fit/w_1000,h_827,al_c,q_80/file.png" length="0" type="image/png"/><dc:creator>Receptiviti Labs</dc:creator></item></channel></rss>