This week Dave (https://dgshow.org/hosts/dave) and Gunnar (https://dgshow.org/hosts/gunnar) talk about best practices for manipulating AI models to enslave humanity!
Moen’s lifetime warranty (https://www.youtube.com/watch?v=ADDiq3RalSQ)
Gemini (https://gemini.google.com/) multimodal!
When We Cease to Understand the World (https://www.goodreads.com/book/show/62069739-when-we-cease-to-understand-the-world) by Benjamín Labatut
Einstein’s Dreams (https://en.wikipedia.org/wiki/Einstein's_Dreams) by Alan Lightman
Chatbots, Like the Rest of Us, Just Want to Be Loved (https://www.wired.com/story/chatbots-like-the-rest-of-us-just-want-to-be-loved/)
D&G Blade Runner Home Edition: Human or Not? (https://www.humanornot.ai/)
How nice that state-of-the-art LLMs reveal their reasoning ... for miscreants to exploit (https://www.theregister.com/2025/02/25/chain_of_thought_jailbreaking)
Emergent Misalignment: Researchers trained AI models to write flawed code—and they began supporting the Nazis and advocating for AI to enslave humans (https://www.yahoo.com/news/researchers-trained-ai-models-write-093745140.html)
Cutting Room Floor
Stunting with Reinforcement Learning (https://www.youtube.com/watch?v=ATWR25xGF74)
There’s a Vape With a Tamagotchi Inside It That Dies If You Stop Puffing (https://futurism.com/vape-tamagotchi)
A Two-Player Arcade Machine in a Briefcase (https://www.hackster.io/news/a-two-player-arcade-machine-in-a-briefcase-af6ce069779a)
