Daniel HuynhinMithril SecurityOur Roadmap for Privacy-First Conversational AIIn September 2023, we released the first version of BlindChat, our confidential Conversational AI. We were delighted with the collective…Oct 10, 2023Oct 10, 2023
Daniel HuynhinMithril SecurityBlindChat: Your Private and Open-Source ChatGPT AlternativeKey takeaways :Sep 27, 2023Sep 27, 2023
Daniel HuynhinMithril SecurityEnhancing AI Transparency and Security with BlindLlama: Open-Source SolutionWe are delighted to announce the launch of BlindLlama, an open-source project that aims to make AI confidential and transparent!Sep 15, 2023Sep 15, 2023
Daniel HuynhinMithril SecurityConfidential Computing: A HistoryIf you haven’t heard about Confidential Computing, here’s the gist. It’s the new buzzword for a trendy tech nobody understands — except for…Aug 3, 2023Aug 3, 2023
Daniel HuynhinMithril SecurityTransforming Enterprises: LLMs Revolutionize AI Deployment for Efficiency and Data SafetyKey Takeaways: Generative AI’s real-world applications are transforming industries by enhancing efficiency and creativity. Deploying AI…Aug 1, 2023Aug 1, 2023
Daniel HuynhinMithril SecurityAI Attacks: Prompt Injection vs. Model Poisoning + MitigationsWith the massive adoption of AI models like GPT, concerns have emerged regarding transparency and safety. Indeed, attackers have performed…Jul 26, 2023Jul 26, 2023
Daniel HuynhinMithril SecurityOpenSource is key for AI transparency but needs more toolingIn a previous , we showed how to hide a malicious model on a model hub to spread false information.Jul 20, 2023Jul 20, 2023
Daniel HuynhinMithril SecurityPoisonGPT: How to poison LLM supply chainon Hugging FaceWe will show in this article how one can surgically modify an open-source model, GPT-J-6B, to make it spread misinformation on a specific…Jul 9, 2023Jul 9, 2023
Daniel HuynhinMithril SecurityThe AI Act: 9 key answers to get onboardThis month, on the 14th of June, the AI Act was successfully passed by the EU parliament. We gathered information on this complex piece of…Jun 29, 2023Jun 29, 2023
Daniel HuynhinMithril SecurityAI-assisted code generation with privacy guaranteesWith the explosion of Large Language Models like ChatGPT, automated code generation, and analysis has well and truly established its role…Jun 28, 2023Jun 28, 2023