Confident Security

Introducing OpenPCC - open-source AI Privacy No tracking. No training. No leaks. Just provably private infrastructure for AI. Check out the repo, announcement, and whitepaper in the comments - hit us with a star while you're there!

blog.confident.security/openpcc

Introducing OpenPCC - open-source AI Privacy No tracking. No training. No leaks. Just provably private infrastructure for AI. Check out the repo, announcement, and whitepaper in the comments - hit us with a star while you're there!

Ever been hustled in a shell game? Imagine a protocol that does that with your sensitive data! We’ve open-sourced our Go implementation of OHTTP (Oblivious HTTP) the protocol that hides who you are from what you send. Link to the repo in the comments, hit us with a star while you're there.

HTTP, but make it binary. This week we've open-sourced bhttp, a Go implementation of RFC 9292. If you need HTTP requests/responses in compact binary form, give it a spin! What do you think we use it for? Repo link in the comments!

Go check out go-nvtrust, our open-source Go library for NVIDIA GPU/NVSwitch confidential attestation. If you've Got problems securing your AI privacy, Go no further, this library will be your Go-to solution. Where are we Going with all of these open-source releases?

Do you care about AI privacy? If so, you'll love our latest open-source release from our private AI stack! Introducing twoway - an encrypted request-response messaging library built on HPKE. Repo and technical breakdown links in the comments

github.com/confidentsec...

Policy’s not gonna save us - Meta’s American Technology Excellence Project will kill any AI privacy legislation they see

Most AI privacy content is noise. The Dead Drop is signal - sharp takes and deep dives from our team for you. First 10 newsletter signups get exclusive privacy swag (ever wanted to block facial recognition?) What AI/privacy topic should we hit?

🛑 People feed sensitive info into AI models every day. Where does it go? Who sees it? Can it be deleted? Without clear answers, your privacy’s at risk. CONFSEC guarantees no visibility, no retention—provably. Watch our CEO talk about #AIPrivacy on TBPN, link on the comments. #Privacy #LLMSecurity

🛑 People feed sensitive info into AI models every day. Where does it go? Who sees it? Can it be deleted? Without clear answers, your privacy’s at risk. CONFSEC guarantees no visibility, no retention—provably. Watch our CEO talk about #AIPrivacy on TBPN, link on the comments. #Privacy #LLMSecurity

This is wild 😳: Meta just fixed a bug that could leak your AI prompts and responses. Turns out our chats weren’t as private as we thought. This is why zero data retention matters. #AI #Meta #Privacy #PromptLeak #LLMSecurity

We loved Helen Nissenbaum's discussion on @techpolicypress.bsky.social's ‬podcast: 🔸Privacy isn’t about hiding—it’s about the appropriate flow of information. 🔸Obfuscation is a response to broken systems, not bad behavior. 🔸When regulation fails, resistance becomes a right. #Integrity #AIPrivacy

𝗡𝗲𝘄 𝗬𝗼𝗿𝗸 𝗷𝘂𝘀𝘁 𝗽𝗮𝘀𝘀𝗲𝗱 𝘁𝗵𝗲 𝗥𝗔𝗜𝗦𝗘 𝗔𝗰𝘁: – 📊 Risk assessments for AI decisions – 🛡️ Transparency + civil rights protections A win for responsible AI implementation. #RAISEAct #InferencePrivacy

@meredithmeredith.bsky.social highlights the real risks of agentic AI: lack of oversight, accountability, and security. ⚠️ Smart agents aren’t necessarily a threat. Unchecked inference is. 🔐 We need private-by-design AI infra. #AgenticAI #InferencePrivacy

📊 𝗠𝗼𝘀𝘁 𝗔𝗜 𝘃𝗲𝗻𝗱𝗼𝗿𝘀 𝗹𝗼𝗴 𝗺𝗲𝘁𝗮𝗱𝗮𝘁𝗮—𝘄𝗵𝗼 𝘆𝗼𝘂 𝗮𝗿𝗲, 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝘀𝗲𝗻𝘁, 𝗮𝗻𝗱 𝘄𝗵𝗲𝗻 𝘆𝗼𝘂 𝘀𝗲𝗻𝘁 𝗶𝘁. Only CONFSEC guarantees that prompts and metadata aren’t stored or visible.

🧠 As AI adoption accelerates, Varonis has looked at how prepared organizations really are for the risks they might bring. They examined 1,000 companies across [healthcare, etc.] and found ghost accounts, weak MFA, and sensitive files exposed. These aren’t new problems—AI just exposes them further.

𝗤𝘂𝗲𝘀𝘁𝗶𝗼𝗻 𝗳𝗼𝗿 𝗔𝗜 𝗯𝘂𝗶𝗹𝗱𝗲𝗿𝘀: 𝘋𝘰𝘦𝘴 𝘴𝘵𝘰𝘳𝘪𝘯𝘨 𝘦𝘷𝘦𝘳𝘺 𝘱𝘳𝘰𝘮𝘱𝘵 & 𝘳𝘦𝘴𝘱𝘰𝘯𝘴𝘦 𝘳𝘦𝘢𝘭𝘭𝘺 𝘩𝘦𝘭𝘱 𝘺𝘰𝘶 𝘤𝘳𝘦𝘢𝘵𝘦 𝘵𝘩𝘦 𝘣𝘦𝘴𝘵 𝘱𝘦𝘳𝘧𝘰𝘳𝘮𝘪𝘯𝘨 𝘮𝘰𝘥𝘦𝘭? 🤔 Follow the thread below ⬇️

𝗪𝗵𝗮𝘁 𝘆𝗼𝘂 𝗻𝗲𝗲𝗱: Model tuning lives on signals, not raw content: latency, token counts, error codes, drift metrics, and opt-in hashed feedback tell you where to improve. 📊

𝗤𝘂𝗲𝘀𝘁𝗶𝗼𝗻 𝗳𝗼𝗿 𝗔𝗜 𝗯𝘂𝗶𝗹𝗱𝗲𝗿𝘀: 𝘋𝘰𝘦𝘴 𝘴𝘵𝘰𝘳𝘪𝘯𝘨 𝘦𝘷𝘦𝘳𝘺 𝘱𝘳𝘰𝘮𝘱𝘵 & 𝘳𝘦𝘴𝘱𝘰𝘯𝘴𝘦 𝘳𝘦𝘢𝘭𝘭𝘺 𝘩𝘦𝘭𝘱 𝘺𝘰𝘶 𝘤𝘳𝘦𝘢𝘵𝘦 𝘵𝘩𝘦 𝘣𝘦𝘴𝘵 𝘱𝘦𝘳𝘧𝘰𝘳𝘮𝘪𝘯𝘨 𝘮𝘰𝘥𝘦𝘭? 🤔 Follow the thread below ⬇️