GoKawiil
Tech News
National Design Studio released Rampart, an open-source system that detects and strips personal information from text before it leaves a user's device. It runs entirely in the browser, combining regex-based rules for structured data like Social Security numbers with the MiniLM machine learning model for names and addresses, aiming to complete the process in milliseconds.
Researchers used cryo-electron microscopy to resolve the structure of a supercomplex formed between nitrogenase and a PII signal transduction protein in a methanogen archaeon. The structure captures how PII, a widely conserved regulator of nitrogen metabolism, physically associates with the nitrogen-fixing enzyme, offering a molecular view of this interaction not previously visualized at this resolution.
Datamimic has released a Community Edition of its data engine as an open-source, MIT-licensed, Python-native tool that generates deterministic synthetic datasets and pseudonymizes PII in staging or QA exports. The CE version runs standalone across databases and file formats like PostgreSQL, MongoDB, CSV and JSON, while a paid Enterprise Platform adds governed workflows, PII scanning, access controls, audit logging and scheduling for regulated industries.
A criminal hacking group has published personal data belonging to nearly 8.7 million customers of Manchester, London Stansted and East Midlands airports after breaching Manchester Airports Group's systems. The attackers had demanded a ransom, which MAG did not pay, and have now released roughly half a terabyte of stolen information—including contact details, vehicle registrations and postcodes—free of charge to other criminals via their own website.
Perplexity has rolled out Hybrid Compute in its Mac app, letting its Computer assistant start tasks in the cloud and shift sensitive steps, like those touching personal files, to a local model running on the Mac itself. An on-device classifier detects personal details such as names and account numbers, replaces them with placeholders before any cloud processing, and restores them once results return. The feature requires Apple silicon, macOS 15 or later, and at least 24GB of unified memory, with 32GB recommended, and uses no cloud credits or API key.