Submissions from mithrilsecurity.io

		Zero-trust AI APIs serving Llama 2 70B inside enclaves (mithrilsecurity.io)
		35 points by DanyWin on Sept 14, 2023 \| past \| 8 comments
		Why open-source is necessary but not sufficient for AI transparency (mithrilsecurity.io)
		1 point by DanyWin on July 20, 2023 \| past
		PoisonGPT: We hid a lobotomized LLM on Hugging Face to spread fake news (mithrilsecurity.io)
		392 points by DanyWin on July 9, 2023 \| past \| 217 comments
		Code generation with privacy guarantees: Deploy SantaCoder with BlindBox (mithrilsecurity.io)
		1 point by DanyWin on July 3, 2023 \| past
		Secret SantaCoder: Confidential AI coding tool protecting code IP (mithrilsecurity.io)
		3 points by Raphael_Millet on June 29, 2023 \| past \| 1 comment
		Audit of BlindAI, an Open-source AI deployment with end-to-end protection (mithrilsecurity.io)
		1 point by DanyWin on May 11, 2023 \| past
		BlindAI API: An open-source and privacy-first OpenAI alternative (mithrilsecurity.io)
		100 points by DanyWin on March 15, 2023 \| past \| 42 comments
		Jupyter Notebooks Are Not Made for Sensitive Data Science Collaboration (mithrilsecurity.io)
		3 points by DanyWin on Jan 26, 2023 \| past
		Jupyter Notebooks Aren't Made for Sensitive Data Science Collaboration (mithrilsecurity.io)
		3 points by Knulp on Jan 26, 2023 \| past
		Build a privacy-friendly AI voice assistant with BlindAI (mithrilsecurity.io)
		2 points by dhuynh95 on May 24, 2022 \| past
		BlindAI, an open-source privacy-friendly AI deployment in Rust (mithrilsecurity.io)
		1 point by lukastyrychtr on May 21, 2022 \| past
		Confidential medical image analysis with Covid-Net and BlindAI (mithrilsecurity.io)
		8 points by dhuynh95 on March 17, 2022 \| past

HN For You