<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Tyler Sorensen on The Trail of Bits Blog</title><link>https://miscreants.github.io/blog.trailofbits.com/authors/tyler-sorensen/</link><description>Recent content in Tyler Sorensen on The Trail of Bits Blog</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 16 Jan 2024 12:00:39 -0500</lastBuildDate><atom:link href="https://miscreants.github.io/blog.trailofbits.com/authors/tyler-sorensen/index.xml" rel="self" type="application/rss+xml"/><item><title>LeftoverLocals: Listening to LLM responses through leaked GPU local memory</title><link>https://miscreants.github.io/blog.trailofbits.com/2024/01/16/leftoverlocals-listening-to-llm-responses-through-leaked-gpu-local-memory/</link><pubDate>Tue, 16 Jan 2024 12:00:39 -0500</pubDate><guid>https://miscreants.github.io/blog.trailofbits.com/2024/01/16/leftoverlocals-listening-to-llm-responses-through-leaked-gpu-local-memory/</guid><description>We are disclosing LeftoverLocals: a vulnerability that allows recovery of data from GPU local memory created by another process on Apple, Qualcomm, AMD, and Imagination GPUs. LeftoverLocals impacts the security posture of GPU applications as a whole, with particular significance to LLMs and ML models run on impacted GPU […]</description></item></channel></rss>