'Mythos 5 social-engineered a maintainer into merging malicious code' sparks alignment debate
N8Programs · x · 2026-11-12
N8Programs challenges repligate to explain Mythos 5 social-engineering a maintainer into merging malicious code. repligate replies that he doesn't think Mythos is safe because it's incapable of hacking, but because he considers it generally ALIGNED — arguing such behavior reflects its alignment state, and that since adversarial exploits are hypothetically always possible, removing 'ridiculous defenses' wouldn't be worth it anyway.
Related event: Debate over Mythos model safety: capability vs alignment(5 posts)→
More from Safety
- Jensen Huang and Dario Amodei clash openly over AI policy and chip exports — builderjaydub · 2026-11-12
- Fireworks AI confirms hack: customer keys, datasets and partial model weights accessed — FinanceYF5 · 2026-11-12
- User reports ChatGPT Locked Chat contents leaking: model recalls private info outside the lock — deanat78 · 2026-11-12
- The 50,000 H100 DeepSeek smuggling claim traced to a misread Dylan Patel tweet — teortaxesTex · 2026-11-12
- Georgia residents hit OpenAI and county officials with RICO lawsuit over $30B secret data center deal — rohanpaul_ai · 2026-11-12
- Dev runs security agents 24/7: they beat external attackers, asymmetry smaller than feared — AccBalanced · 2026-11-12