CipherExplain
SHAP explanations on encrypted data — vendor never sees it
What it does
CipherExplain runs SHAP model explanations on encrypted inputs using fully homomorphic encryption. The vendor never sees plaintext features, model weights, or explanations. Built for regulated AI buyers (SR 11-7, EU AI Act, HIPAA). Python SDK.
Does the same job
all alternatives →- S2SHA-256 explained step-by-step visually2022 · sha256algorithm.com · ▲1,241
- RWRead Wikipedia privately using homomorphic encryption2022 · spiralwiki.com · ▲331
Hi, creator here. This is a demo of our recent work presented at Oakland (IEEE S&P): https://eprint.iacr.org/2022/368. The server and client code are written in Rust and available here: https://github.com/menonsamir/spiral-rs. The general aim of our work is to show that homomorphic encryption is practical today for real-world applications. The server we use to serve this costs $35/month! A quick overview: the client uses homomorphic encryption to encrypt the article number that they would like to retrieve. The server processes the query and…
- BABuilding an End-to-End Encrypted Shazam with Homomorphic Encryption2024 · zama.ai · ▲59
- AAAriadne – A Rust implementation of aperiodic cryptography2025 · codeberg.org · ▲40
Hello HN, we're CipherNomad, the research initiative behind this project. The Ariadne Protocol is our exploration of a different cryptographic model. The work began with an observation of primitives like the Lion transform, which use a static, hardcoded sequence of operations. This led us to ask: What if the cryptographic "program" wasn't a constant, but a dynamic, history-dependent variable? Our first step was a "Cryptographic Virtual Machine" that took an explicit list of operations (a "Path"). This worked, but required sharing the Path object—an explicit dependency that needed to be…
- LLLightPHE: Lightweight Partially Homomorphic Encryption for Python2025 · github.com · ▲6
Homomorphic encryption enables computations to be performed directly on encrypted data without requiring access to the private key. This allows data to be securely stored in the cloud while still being processed using the cloud’s computational power. Meanwhile, the cloud remains unaware of the actual data it is handling. Although fully homomorphic encryption has become available in recent years, when considering performance trade-offs, partially homomorphic encryption proves to be a more efficient and practical choice. It is significantly faster, requires fewer computational resources, and…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, May 2026
the whole month →
Parallel agents, diff reviewer, and multi-model comparisons
Dev tools · May 2026 · kilo.ai
- NW
Hey HN, Henry here from Cactus. We open-sourced Needle, a 26M parameter function-calling (tool use) model. It runs at 6000 tok/s prefill and 1200 tok/s decode on consumer devices. We were always frustrated by the little effort made towards building agentic models that run on budget phones, so we conducted investigations that led to an observation: agentic experiences are built upon tool calling, and massive models are overkill for it. Tool calling is fundamentally retrieval-and-assembly (match query to tool name, extract argument values, emit JSON), not reasoning. Cross-attention…
Life & fun · May 2026 · github.com
- FM
Dev tools · May 2026 · github.com