HE CNN inference paper: channel-interleaved packing and a small rotation-key set, trimming the usual encrypted-conv tax.

Can TFHE bootstrapping get cheaper by splitting lookup tables and doing homomorphic div/rem?

Can federated learning keep its accuracy once you bolt on HE and DP, or does client heterogeneity still wreck it?

Sublinear encrypted comparisons for PNN search, by clustering and homomorphic PQ, because fully homomorphic pain needed pruning.

Boolean functions get partitioned by inner products, then analyzed for slice complexity, GSW-style homomorphic eval, and crypto metrics.

Attested TLS can be relayed, so confidential computing’s data sovereignty pitch takes a CVE-sized hit.

FHE for AI trading, crypto finance, and tokenized assets, with the usual latency and infra bill left as an exercise.

A zkML explainer on proving inference ran correctly, with circuits, quantization, and the usual proof-generation pain.

Paper on LibFHE, a Numba/CUDA-Python GPU FHE library for non-RNS CKKS-BGV, because C++ was too mainstream.

Paper on private, robust, verifiable FL aggregation with multi-key FHE for Krum, Trimmed Mean, and FLTrust, because one trust model was t...

Non-RNS CKKS-BGV on GPUs, because apparently someone missed the usual RNS memo and still got competitive performance.

Can you fine-tune a model privately with off-the-shelf MPC/HE and still benchmark it on a real PPML task? PriFT says yes.

Walsh LUTs on lazy CKKS bits speed up AES transciphering, because apparently XORs deserved a detour.

Zero-knowledge, FHE, MPC, and TEEs are now product infra in 2026, so the real choice is which trust model you can stomach.

Non-RNS CKKS-BGV on GPUs, apparently via Numba, and the paper claims C++-library-level performance without the C++.

PSOs at PSI speed, by swapping curve tricks for RLWE OKVS decoding and batching the whole mess.

Paper on prefix scans for noncommutative monoids in bit-reversed HE slots, cutting rotations to the theoretical minimum.

YouTube talk on computing over encrypted data, i.e. the usual FHE dream with extra ciphertext overhead.

Can SNLP cut FHE Transformer depth? This paper says yes, with fewer bootstraps and only modest perplexity loss.

HEAD-FL: federated learning with round-adaptive DP and verifiable homomorphic aggregation, because fixed noise was too honest.

Generalizes CKKS key-switching decomposition to many inputs/keys, so batched rotations/relinearizations get less noise and less waste.

RotPIR paper on sublinear single-server PIR with FHE preprocessing via oblivious rotation, because one server was too easy.

Weekly privacy-news archive from Web3Privacy Now, with ZK/FHE/MPC/messaging/policy updates, because privacy never sleeps.

NIST's PQC migration FAQ covers threat timing, crypto inventories, risk, planning, testing, and validation.

What would it actually take to make iO practical, and why are FHE, ZKP, and MPC people all involved?

The hard part is versioned ciphertext, this Microsoft post-quantum crypto-agility post covers JOSE, CMS, Office, SQL Server, Blob Storage.

Fixes FLPSI's soundness bug by making the homomorphic matching composable, so false accepts don't pile up with database size.

What’s Arcium doing? A live dashboard of encrypted compute, program execution, and throughput metrics.

zkSecurity's zkao v1 is a deep-research bug finder for crypto codebases, still KYB-gated because apparently scanners need adult supervision.

PCI SSC interview with Utimaco on HSMs, key management, AI crypto risk, and PQ migration, because governance needed another acronym.