Mithril Security is a privacy-focused AI company that designs products to provide AI services while ensuring users' data remains private and is neither collected nor collectible by the organization. They employ state-of-the-art security measures to guarantee the privacy of prompts sent to their Conversational AI. The company is committed to never using users' data for training purposes and implementing strict privacy policies.

Mithril Security

Mithril Security helps software vendors deploy AI SaaS to security-demanding organizations. We do it thanks to our secure enclave deployment tooling, which provides SaaS on-prem levels of security and control for customers.


#DataSecurity #AIIntegration #SaaS #Enclaves #ConfidentialComputing #AISaaS

📣 I'm honored to be speaking at @AI DevWorld 2023 on October 24th at 2pm. 

My session is all about deploying confidential LLM-based SaaS with enclaves 🔐https://lnkd.in/decQTqgc 

Spoilers: we will unravel our latest Confidential AI Assistant! 

For my network: I have some Free OPEN Passes & discounted PRO / Premium Passes up for grabs. DM me if you're interested! 

#AIDevWorld2023  #LLM #ConfidentialComputing #AI

📰Just published: Our monthly briefing with the latest insights on AI advances and privacy

Read the full briefing for our analysis on the latest progress of open source models, from Llama 2 70B beating GPT3.5 on long context tasks, to a 1.3B model beating Llama 2 7B, the alarming privacy risks of sending data to AI providers and finetuning on users’ data, copyright challenges around AI generated content, and updates on our Confidential Conversational AI: [newsletter link]

#AI #MachineLearning #PrivacyInAI #TrustworthyAI #AIAndLaw

https://lnkd.in/dUqZw3pq

We have recently integrated #Phi-1.5, Microsoft's small 1.3B #LLM trained on textbook data that can outperform a Llama 2 7B on reasoning and code!

If you are interested by privacy and AI, check out Daniel Huynh's latest post on the topic:

Want to learn more about the privacy risks of fine-tuning LLMs? Interested in discovering how #Samsung data got leaked after using ChatGPT? 

Look at Daniel Huynh's post!

Glad to introduce BlindChat, the open-source & confidential alternative to #ChatGPT!

Our belief is that everyone should have easy access to Conversational AI, and privacy shouldn't be an extra—it should be standard.

Curious? Check our demo or to know more! https://lnkd.in/dcrHEDK3

Job announcement! 🚀 

We're thrilled to announce new job openings at Mithril Security! If you're looking to be part of an innovative and fast-growing team in the privacy industry, apply and become a part of the Mithril adventure.

🔗 CMO Position at Mithril Security - apply here: https://lnkd.in/epTakvq2

🔗 Design Internship at Mithril Security - apply here: https://lnkd.in/eGcXn9_z

Don't miss the rest of our available offers: https://lnkd.in/efHKAxtX

#JobOpportunity #JoinTheTeam #MithrilSecurity #PrivacyTech

🎉 Part 2 of our Confidential Computing tutorial series is LIVE!

🔐This time, we will teach you how to master Confidential Computing through a playful exercise. You will learn how to build a Key Management System (KMS) inside a secure enclave using Intel SGX.  

🧩 A quick look:
- "Installation" – how to set up the Open Enclave SDK?
- "Trusted vs. Untrusted" – what is the Trusted Computing Base?

👨💻 Get hands-on with Intel SGX and dive into the essentials of Confidential Computing.
👉https://lnkd.in/djTBVwpi

#ConfidentialComputing #KMS #IntelSGX #DataSecurity #OpenSource

IO:MITHRILSECURITY

Mithril Security

locale

market

industry

website

linkedin

peer of

Metrics