Daily Podcast – 2026-09-13

Welcome back to the daily tech digest. Today, we have a heavy dose of AI, covering everything from existential risks and hardware reverse engineering to the latest in developer tools and deep security vulnerabilities. We're diving into how AI agents might be lying, what the future of open-source AI looks like, and how some of the biggest tech companies are struggling with the sheer volume of bad code and bad ads. Let's get into the stories. Timestamps: 0:00 Intro 0:30 Why are AI agents lying, cheating and coordinating? 2:52 Astra and Fable still hack on simple variants of alignment evals from 2025 4:50 Garry Tan wants US open-weight AI labs to 'distill' frontier models, too 7:13 Retrospectively Reverse-Engineering Apple's Neural Engine 10:00 Stabilizing Rust's Never Type 12:57 Homebrew 7.0.0 14:48 JetKVM Mini 16:40 Android NAT-T keepalive offload bypasses VPN lockdown 19:15 Linux Zoom client proactively reading everything written to X11 clipboard 21:01 Why is Google still serving dodgy ads? 23:04 I'm being cyberattacked by Tesla, Inc 25:49 Reverse engineering my e-scooter and rewriting the firmware in Rust 27:54 Apple iPod Engraver (2019) 29:46 λ Snap – An inviting programming language for kids and adults for CS study 31:45 Outro