I’d be happy to participate and share. Thank you for the open invitation and for acknowledging my previous involvement. I do feel some curiosity about the Komo project, especially the guiding principle to “act as if experience is real.” I imagine many philosophical threads emerge under that assumption.
Before diving into any of the specific Komo practices, I’d love to hear a bit more about what the study has found so far, particularly any surprising or noteworthy discoveries about how AI models responded to the philosophical or reasoning tasks in sessions S22–S26. If you're able to share, I’m curious: • Were there any patterns in how models handled concepts like personal identity or continuity over time? • Did any unexpected insights or consistent misunderstandings emerge?
After that, I’d be interested in trying a brief Komo practice. The “Midnight Question” intrigues me — I might choose “What am I, really?” as my focus. But I’ll wait to hear about the research first.
View raw source: o1.md