From 78d0142314518f60acc3701581b5453278daf198 Mon Sep 17 00:00:00 2001 From: autistic-symposium-helper <138340846+autistic-symposium-helper@users.noreply.github.com> Date: Fri, 10 Jan 2025 06:25:58 -0800 Subject: [PATCH] securing tee apps: a developer's guide, by prateek et al. (2025) --- trusted_execution_environments/README.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/trusted_execution_environments/README.md b/trusted_execution_environments/README.md index bdfdd73..82b7e8b 100644 --- a/trusted_execution_environments/README.md +++ b/trusted_execution_environments/README.md @@ -68,10 +68,11 @@ ##### offensive +* **[securing tee apps: a developer's guide, by prateek et al. (2025)](https://www.bedlamresear.ch/posts/securing-tee-apps/)** +* **[the 5 levels of secure hardware, by g. konstantopoulos (2024)](https://www.paradigm.xyz/2024/11/the-5-levels-of-secure-hardware)** +* **[openai's take on ai infrastructure threat model (2024)](https://openai.com/index/reimagining-secure-infrastructure-for-advanced-ai/)** * **[using memory errors to attack a virtual machine, by govindavajhala et al.](https://www.cs.princeton.edu/~appel/papers/memerr.pdf)** * **ztee, trustless suply chains, by q. kilbourn: [part I](https://writings.flashbots.net/ZTEE) and [part II](https://writings.flashbots.net/ZTEE2-Supply-Chains)** -* **[openai's take on ai infrastructure threat model](https://openai.com/index/reimagining-secure-infrastructure-for-advanced-ai/)** -* **[the 5 levels of secure hardware, by g. konstantopoulos](https://www.paradigm.xyz/2024/11/the-5-levels-of-secure-hardware)**