Final Notice – Free Online Workshop: Visualizing Corrigibility, Alignment, and Safety (Aug 10)

[Andy E Williams]

You’re invited to a free online workshop:

“Visualizing Corrigibility, Alignment, and Safety from First Principles”

🗓 Date: August 10, 2025

⏰ Time: 9:00 am, Time Zone: Reykjavik, Iceland.

🌐 Location: Online (Zoom link provided upon registration)

🔗 Register: https://events.payqlick.com/event/51/AGI%20Conference%202025

This workshop offers a visual and formal walkthrough of the Functional Model of Intelligence (FMI)—a first-principles framework where corrigibility, alignment, and safety are built-in geometric properties of conceptual and cognitive fitness spaces, stable even under recursive change.

Included in this session:

• A participant poll on foundational alignment intuitions

• Formal definitions of intelligence, corrigibility, and safety as recursive properties

• New visualizations of alignment as structural convergence in concept space

• Introduction of the Recursive Alignment Infrastructure Platform (RAIP)

• Live discussion of example simulations, including:

  • Predicted collapse in the coherence of alignment research without FMI

  • Phase transition in collective reasoning under RAIP

  • Conceptual space instantiation via finite primitives
    

🧩 Download Full Workshop Notes (with diagrams): https://drive.google.com/file/d/1JEZD8e4B6iklrXulnEbNQrdyHtYoiTSI/view?usp=drive_link

Whether you're a theorist, engineer, or alignment researcher, this workshop is designed to let you see alignment failure and recursive safety—not just debate it.

Poll (2 minutes):
On attending, we invite you to take a brief poll about the necessity of a functional model of intelligence in achieving AI alignment. This poll will be distributed upon registration.

Results will be reviewed live.

Looking forward to your participation,

Caribbean Center for Collective Intelligence