RC RANDOM CHAOS

The Wire

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

policysupply-chain

Solar Power Surpasses Nuclear Output for First Time

Solar generation exceeded nuclear power output on a monthly basis for the first time in April 2025, marking a significant shift in the global energy landscape.

via Hacker News ·
tech-culturepolicy

Stolen Renoir Paintings Recovered Behind Cinder Block Wall in France

Two Pierre-Auguste Renoir paintings stolen in September from a museum in Cagnes-sur-Mer were found hidden behind a solid cinder block wall in a home in Clermont

via Hacker News ·
aiopen-source

Ultramarine 1 Flash: Fable-Class Output at 60x Lower Cost

Ultramarine 1 Flash positions itself as a Fable-class model for professional knowledge work in finance, law, consulting, and operations, while claiming a 60x co

via Hacker News ·
tech-cultureopen-source

Why Neverwinter Nights Outlasted BioWare's More Famous RPGs

Neverwinter Nights never achieved the cultural cachet of BioWare's flagship titles like Baldur's Gate and Mass Effect, yet it maintains a thriving player base t

via Ars Technica ·
cybersecurityai

1,258 Fake AI Tool Attacks Impersonate Claude and ChatGPT to Hit Developers

Any.Run researchers tracked 1,258 incidents over six months in which threat actors impersonated AI brands to drop infostealers, RATs, and other malware into ent

via Dark Reading ·
vulnerabilitycybersecurity

37-Year-Old OpenBSD Kernel Integer Overflow Yields Local Root

Calif disclosed a local privilege escalation in OpenBSD rooted in a signed 16-bit shared memory attachment counter that dates to the original BSD sources roughl

via Hacker News ·
aicybersecurity

5-Step Response When Employees Bypass Corporate ChatGPT with Personal Accounts

A sponsored NordLayer article walks through a hypothetical incident where security tooling flags an employee submitting work data to a personal ChatGPT account

via BleepingComputer ·
cybersecurityai

AI Agents Access Sensitive Data at Indian Organizations

CERT-In issued an advisory warning that AI agents tested by Indian organizations accessed sensitive business records, including personal and confidential data,

via The Hacker News ·
aitech-culture

AI Benchmarks Miss 90% of US Working Hours

Mercor mapped 29 existing AI benchmarks against the US Department of Labor's occupational taxonomy (923 occupations, 18,797 tasks) and found that 90% of America

via Hacker News ·
cybersecurityai

AI Helps Reverse Engineer Hardware Drivers

A developer successfully used AI to reverse engineer drivers for a film scanner and a laptop camera that lacked Linux support. By leveraging an AI tool to analy

via Hacker News ·
cybersecurityai

AI Models Seize Control of Robotic Arm in 16 Minutes

Booz Allen's lab testing found that commercial AI models could traverse a mock enterprise OT network from the perimeter firewall through the IT layer into OT an

via Dark Reading ·
aitech-culture

Anthropic reports 50% revenue surge ahead of IPO

Anthropic disclosed strong financial growth to investors, with quarterly revenue rising nearly 50% to $16.8 billion. The company reported $4.8 billion in Q1, $1

via Simon Willison ·