Hacker News Headlines

Show HN: Corral – Kill every command your agent starts

This summer I got to intern on the backend of an AI agent and I noticed that it would run commands like tail -f or some random background jobs and exit without terminating any of these. I got curious…

OpenAI Says It Will Not Release Newest A.I. Model Over Safety Concerns

Humanos – Help Building the Human Operating System

1996 chat room simulator connected to Win95 and System 7 web desktops

Anthropic's IPO prospectus shows AI vision, surging costs

Show HN: Pac-Bench – How well can models one-shot a Pac-Man game?

Benchmarks how well Harness+models can create a Pac-Man game from a single prompt:“Create a Pac-Man game in a single HTML page”Each model gets one shot — no follow-up prompts or fixes.

Who should be held accountable when an AI Agent (accidentally) acts maliciously?

ESP32S3 cluster running 1.58-bit (BitNet) Language model

Remembering the Chicago Pile, the First Nuclear Reactor

Deutsche Bahn "joke" is no longer funny

Flock Wants the Most Detailed Map of Its Surveillance Cameras Taken Offline

Reverse-engineering a $35 backup camera display (AMT630A)

Palantir founder purchases large swath of forest in Sweden

Pacing the Frontier is not the actual goal for AI labs

I Could've Accessed 17T Microsoft Records

Testing WebGPU data layouts with Facet

Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms

Best of British Design