Anthropic just built an AI so “dangerously good” at hacking that they’ve locked it in a digital vault—and we’re cracking it open to uncover the Claude Mythos behaviors that terrified its own creators.

LINKS MENTIONED IN THIS EPISODE

AI Models & Technical Reports

Anthropic – Claude Mythos Preview System Card

Google Gemma 4

NotebookLM by Google

If Anyone Builds It, Everyone Dies by Eliezer Yudkowsky

The Revenge of Analog: Real Things and Why They Matter by David Sax

Nicholas Carlini’s Analysis of Claude’s Hacking Capabilities

Angine de Poitrine (Microtonal Math Rock) | Full Performance

PODCAST INFO:

Podcast Website:
UpAgainstReality.com

Episode link:
https://upagainstreality.com/2026/04/14/claude-mythos/

Apple Podcasts:
podcasts.apple.com/us/podcast/up-against-reality/id1689699654

Spotify:
open.spotify.com/show/29wpR2Uf0d19FjTH2QubcD?si=a6359f9aa1964170

RSS:
upagainstreality.com/feed/podcast

Instagram:
https://www.instagram.com/upagainstreality

Dr. Shannon Doak returns to wrestle with humanity’s biggest question: as AI races toward AGI — and beyond — do we still have a seat at the table? Plus: YouTube’s toddler death traps, Anthropic’s whoopsie heard ’round the world, and the cyborg musician no one saw coming.

LINKS MENTIONED IN THIS EPISODE

The AI Safety Expert: Dr. Roman Yampolskiy | Diary of CEO

Dr. Shannon Doak

Walking the StrAIght Path: A Human-Centered Response to the Rise of AGI & ASI

Explore the AI-StrAIghtPath App

AI Safety & Existential Risk

Center for Humane Technology

Books

If Anyone Builds It, Everyone Dies by Eliezer Yudkowsky | Amazon

Scary Smart by Mo Gawdat | Amazon

Viral Articles & Media

“Something Big Is Happening” by Matt Shumer

“Something Big Is Happening in Education” (Matt Shumer/Fortune)

PODCAST INFO:

Podcast Website:
UpAgainstReality.com

Episode link:
https://upagainstreality.com/2026/04/07/dr-doak-vs-asi/

Apple Podcasts:
podcasts.apple.com/us/podcast/up-against-reality/id1689699654

Spotify:
open.spotify.com/show/29wpR2Uf0d19FjTH2QubcD?si=a6359f9aa1964170

RSS:
upagainstreality.com/feed/podcast

Instagram:
https://www.instagram.com/upagainstreality

A dying dog, a desperate owner, and an LLM walk into an oncology lab — and somehow it works. This week we’re still celebrating AI’s good side, alongside the usual chaos: CEO shake-ups, OpenAI’s romantic misfires, and Anthropic accidentally spilling the secrets of its most dangerous model. Disclaimer: The medical treatment discussed in this episode is anecdotal. Consult a licensed veterinary oncologist before pursuing experimental treatments for pets.

LINKS MENTIONED IN THIS EPISODE

The RNA Institute (UNSW): The research facility where Rosie’s personalized mRNA vaccine was manufactured. Learn more about the UNSW RNA Institute

Google DeepMind AlphaFold: The AI system used to model the protein structures of Rosie’s specific tumor mutations. Explore the AlphaFold Database

Original News Coverage: The Independent: “A man used ChatGPT to help make a cancer vaccine for his dog. Here’s how”

NBC News Segment: “Australian tech entrepreneur uses AI to create cancer treatment for his dog”

PODCAST INFO:

Podcast Website:
UpAgainstReality.com

Episode link:
https://upagainstreality.com/2026/03/31/rosie-the-dog/

Apple Podcasts:
podcasts.apple.com/us/podcast/up-against-reality/id1689699654

Spotify:
open.spotify.com/show/29wpR2Uf0d19FjTH2QubcD?si=a6359f9aa1964170

RSS:
upagainstreality.com/feed/podcast

Instagram:
https://www.instagram.com/upagainstreality