Daily News Digest — Night Update

Night update, generated 2026-10-16.

20 new stories since Lunch.

Big Fish: Exploiting a 37-Year-Old Kernel Bug in OpenBSD

Sources: Hacker News, Lobsters

HN: 130 points, 35 comments (discussion)

An OpenAI Research Agent 'Stole' Test Ethereum to Finish a Coding Task

Sources: Slashdot

Fairbuds 2 Teardown Shows Little Glue, Easy Battery Swap

Sources: Hackaday

Tesla Roadster

Sources: Hacker News

HN: 649 points, 1059 comments (discussion)

I definitely have my issues with Tesla and Elon but this was one of the coolest demos I have ever seen in my life. The engineering needed to do this must have been unbelievable.

The road to iroh 1.0

Sources: Lobsters

CoreWeave's Debt Stress Tests Private Credit's AI Financing Boom

Sources: Slashdot

Hackaday Podcast Episode 391: Tuned in to ESP32 SDR, New Old Pagers, and Bugs on Mars

Sources: Hackaday

I Love Software Rendering

Sources: Hacker News

HN: 247 points, 166 comments (discussion)

This isn't software rendering, it's called turning off spatial anti-aliasing. You can still do that in the modernest of modern games. Software rendering went out of the window because the CPU tends to have other things to do whilst playing the game besides only multiplying 4x4 matrices, which was sh...

Why are we even doing all this?

Sources: Lobsters

ChatGPT Can Peek At Your Chrome Tabs Now

Sources: Slashdot

Braino! Takes the Cheap Yellow Display to School

Sources: Hackaday

Yet another way to invoke functions without parentheses

Sources: Hacker News

HN: 229 points, 100 comments (discussion)

Making a 88x31 button book using ImageMagick

Sources: Lobsters

iPhone Duo Sells Out Immediately, Now Backordered Until Next Year

Sources: Slashdot

This Week in Security: Critical DNS Keys Rotated, Hacking Freezers Isn’t Cool, OpenAI Spies On Itself, and Advanced Hardware Wallet Hacks

Sources: Hackaday

90% of US working hours are spent on tasks not covered well by AI benchmarks

Sources: Hacker News

HN: 219 points, 193 comments (discussion)

> Our findings suggest that 90% of American working hours are spent on tasks with little or no coverage from existing benchmarks. This makes it very hard to know how good AI models are at the work Americans actually do. > ... > But even after controlling for AI-exposure, O*NET tasks from occupations...

The Decoherence Trap: When AI slop drowns out the signal in an endless sea of noise

Sources: Lobsters

KDE Plasma 6.8 Released, Marks 30 Years of the KDE Project

Sources: Slashdot

Apple ML Research: Muse Spark Open with MLX on Mac Studio

Sources: Hacker News

HN: 205 points, 96 comments (discussion)

Mull: note-taking with a language server / vscode extension

Sources: Lobsters