Writing

• Ideonomy
  • Introduction to Ideonomy
  • Properties and Dimensions in Ideonomy
  • Negation in Ideonomy
  • Sensory Transducers
• Machine Behavior and Human-AI Interaction
  • Novel Uses For LLMs
  • Advantages of AI Therapists
  • How might LLMs detect injected tokens? (Leaflet)
  • Distinguishing between LLM injection detection methods (Leaflet)
• AI Alignment
  • A Three-Facet Framework for AI Alignment
  • Subtypes of Control in AI Alignment
  • A Simple Evaluation for Transformative AI
  • Can Mechanistic Interpretability Help With Prompt Injection?
• Miscellaneous
  • The Unbearable Weight of the Wayback Machine
  • Preparing the World for Superintelligence
  • Revisiting Patrick Gunkel's 'Future Headlines'
  • The Anti-Preparedness Paradox
  • X-Risks and S-Risks From Alien Invasion
  • How to Think

Note: I mostly write new posts on my blog these days.