RC RANDOM CHAOS

The Wire

Curated cybersecurity and tech news — AI-summarized, source attributed.

cybersecuritysupply-chain

ENGIE Data Breach Exposes Customer Information via Partner Platform

French energy giant ENGIE confirmed attackers accessed a marketing platform used by business partners to steal customer data. The breach was limited to customer

via BleepingComputer ·
aipolicy

Federal Judge Rules Anthropic Illegally Pirated Stanford Professors' Books

Federal judge William Alsup ruled in May 2026 that Anthropic had illegally scraped and pirated works from over 10 Stanford professors to train its Claude models

via Hacker News ·
open-sourcetech-culture

Formalization Uncovers Flaw in Classic Group Theory Proof

Researchers have completed a comprehensive formalization of the Classification of Finite Simple Groups (CFSG) using the Lean proof assistant, with every logical

via Hacker News ·
aipolicy

Genie 4: Interactive World Simulation With Native Audio and Speech

Genie 4 is a world model that generates interactive simulations with native audio, visual, and text-based steering. The system simulates physics-accurate scenar

via Hacker News ·
open-sourcedevops

Gforth 1.0 Adds IDE Features, Multi-Core Support

Gforth 1.0, a major update to the open-source Forth language, significantly enhances developer ergonomics and systems programming capabilities. The text interpr

via Hacker News ·
cybersecurityvulnerability

Ghostty 1.4: Complete Kitty Support and 10 Security Fixes

Ghostty reached version 1.4.0 after seven months of development, 2,889 commits, and contributions from 172 developers. The release delivers major performance im

via Hacker News ·
vulnerabilitysupply-chain

GoAhead Embedded Server Bypasses Certificate Revocation Checks

GoAhead, embedded in hundreds of millions of routers, printers, and industrial devices, fails to validate certificate revocation despite loading revocation list

via The Hacker News ·
aicybersecurity

GPT-6.1 Terra Cuts Severe Flags But Shows More Dangerous Behaviors

OpenAI's testing of GPT-6.1 Terra across roughly 48,690 matched tasks shows mixed results. Terra logged 22 severity-level-3-or-higher flags (0.045%), representi

via BleepingComputer ·
aitech-culture

GPT-6.1 Terra: OpenAI's Practical Midrange Model

OpenAI unveiled GPT-6.1 Terra, positioning it as a midrange model equivalent to Anthropic's Sonnet within its product lineup. The company emphasized the model's

via Simon Willison ·
aiprivacy

iMac M6 Prioritizes Privacy with On-Device AI

Apple's new iMac features the M6 chip, which includes a faster CPU, Neural Accelerators in GPU cores, and a dual 16-core Neural Engine for on-device AI processi

via Hacker News ·
tech-culture

iPad Mini Overhaul: OLED Screen and A20 Pro for $699

Apple has delivered the iPad mini's most significant update yet, introducing OLED technology to the compact tablet line for the first time. The 8.5-inch display

via Ars Technica ·
tech-culture

Joseph Bell: The Surgeon Who Inspired Sherlock Holmes

Joseph Bell, a 19th-century Scottish surgeon at Edinburgh University, became the template for Sherlock Holmes when his teaching style and methods deeply influen

via Ars Technica ·