#privacy News & Analysis

New ways to buy ChatGPT ads

Starknet Phase 4 Is Here: Shinobi Upgrade Brings Native Privacy and Bitcoin Support to Mainnet

Starknet has launched its Shinobi upgrade on Mainnet, introducing protocol-level native privacy and Bitcoin integration through strkBTC. The upgrade marks Phase 4 of Starknet's development roadmap, with eight new projects deployed including Privily, a privacy-focused L3 neobank, while the ecosystem celebrates Loot Survivor reaching 100 million onchain transactions.

Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents

Enabling privacy-preserving AI training on everyday devices

CoreGuard: Safeguarding Foundational Capabilities of LLMs Against Model Stealing in Edge Deployment

Startale taps Privacy Boost to bring self-custodial privacy to Sony-backed Soneium app

Litecoin rewrites three hours of history to undo its first major privacy-layer exploit

Separable Expert Architecture: Toward Privacy-Preserving LLM Personalization via Composable Adapters and Deletable User Proxies

Federated Co-tuning Framework for Large and Small Language Models

Yuval Rooz: Achieving on-chain capital markets requires solving privacy and settlement finality, misconceptions about Canton are unfounded, and super validators play a crucial role in transaction processing | Unchained

The question isn't whether privacy. It's what sort of privacy

Privacy protocol Umbra shuts front end to stifle Kelp exploiters

An AI Agent Execution Environment to Safeguard User Data

Introducing OpenAI Privacy Filter

Coinbase Study Finds Privacy Tools Like Railgun Are Mathematically Safe From Quantum Attacks

Toward Zero-Egress Psychiatric AI: On-Device LLM Deployment for Privacy-Preserving Mental Health Decision Support

Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs

Rethinking Post-Unlearning Behavior of Large Vision-Language Models

Distributed Human Identity: AI-Enabled Multi-Existence Through Cognitive Replication and Robotic Embodiments

Privacy-Preserving LLMs Routing

Researchers propose PPRoute, a privacy-preserving framework for LLM routing that uses Secure Multi-Party Computation (MPC) to protect user data while dynamically selecting between model providers. The system achieves 20x speedup over naive MPC implementations through optimized encoder inference, multi-step model training, and an efficient Top-k algorithm, maintaining routing quality without sacrificing privacy.

Synthetic data in cryptocurrencies using generative models

Researchers propose using Conditional Generative Adversarial Networks (CGANs) to generate synthetic cryptocurrency price data, addressing privacy and access concerns in financial research. The approach combines LSTM generators with MLP discriminators to produce statistically consistent synthetic time series that preserve market dynamics, offering a computationally efficient alternative for financial modeling and analysis.

Tempo’s ‘Zones’ sparks debate over privacy in crypto infrastructure

Tempo has launched 'Zones,' a feature enabling private, permissioned stablecoin transactions on its layer-1 blockchain targeting enterprises. The development has triggered criticism from the crypto community, with detractors arguing that the operator-controlled architecture reintroduces centralized trust mechanisms, potentially undermining core decentralization principles.

Blockchain Is South Korea’s New Fiscal Weapon — A Blow To Privacy?

Tempo Unveils 'Zones' for Private Enterprise Stablecoin Transactions

From Anchors to Supervision: Memory-Graph Guided Corpus-Free Unlearning for Large Language Models