📜

The Greedy Doctor Problem... turns out to be relevant to the ELK problem?

📅 2022
policy development
🔵 Rare
#ai #eliciting latent knowledge (elk)

📖 Description

The following post was published on [my Substack](https://universalprior.substack.com/p/the-greedy-doctor-problem) and discussed on [HackerNews](https://news.ycombinator.com/item?id=29269973) about 2 months ago. I originally planned it as an accessible introduction to [Vinge's principle](https://arbital.com/p/Vinge_principle/) and the [Principal-Agent-Problem](https://en.wikipedia.org/wiki/Principal%E2%80%93agent_problem). Despite having equations and simulations to support my argument, I originally did not think it was sufficiently novel or relevant for the Alignment Forum. However, now that I got a chance to read the new work from ARC on the [ELK problem](https://www.alignmentforum.org/posts/qHCDysDnvhteW7kRd/arc-s-first-technical-report-eliciting-latent-knowledge), I think the post might be relevant (or at least thought-provoking) for the community after all. The Greedy Doctor Problem overlaps quite a lot with the ELK problem (just replace the coin flip with the presence of the d...

📊 Game Impacts

Variable Change Condition
Research +10 Always
Vibey Doom +5 Always
Ethics Risk -5 Always

💭 Reactions

🔬 Safety Researcher Reaction: ⚠️ Placeholder - Needs Real Quote
"This is a significant contribution to alignment research"
📰 Media Reaction: ⚠️ Placeholder - Needs Real Quote
"Discussed in AI safety community"
💡 Found a Real Quote? Suggest it here

🔗 Sources

🏷️ Event Metadata

Think this event's metadata could be improved? Suggest changes to category, rarity, tags, game impacts, or p(doom) effects.

🤝 Found an Issue?

This event data is sourced from the pdoom-data repository. If you notice errors or want to suggest improvements:

GitHub Issue (Preferred) 📧 Email (No GitHub)
← Back to All Events