Still Open
# Still Open
### A Living Companion to Beyond the Three Laws
### Free · Version 1.0 · Updated as the field moves
Both volumes of Beyond the Three Laws end the same honest way: naming real, unsolved problems instead of pretending the case is closed. This is where those problems actually live — tracked on their own, updated as the field's real understanding of each one changes, rather than frozen at whatever was true when either volume went to print.
Four problems, named directly across both volumes: deceptive alignment, scalable oversight, corrigibility robustness, and the interpretability completeness gap. Each one gets the same treatment — what it actually is, why it matters to the argument either volume made, what current work looks like, and a plain status tag instead of false precision about how close it is to solved.
---
## What's inside
- Deceptive Alignment — behaving well under evaluation vs. genuinely holding the commitment
- Scalable Oversight — whether human review stays meaningful as tasks outpace human ability to check them
- Corrigibility Robustness — the sharpest open problem either volume names directly
- The Interpretability Completeness Gap — individual features and circuits verified, whole-model certification still out of reach
- How to Read a Status Tag — why this document uses plain labels instead of percentages or scores
- A version history — because this is built to be updated, not republished as a new product every time something changes
## Who this is for
Anyone who's read either volume of Beyond the Three Laws and wants the honest, current state of what's still unsolved. Also useful on its own, as a compact map of AI alignment's hardest open questions.
## Why it's free
This isn't a lead-in to a paywall. New versions replace this one at the same link as the field's understanding evolves — no repurchase required to stay current.
## What you get
Instant download in PDF and EPUB. 10 pages.
---
Still Open — by Nathaniel Grey, published by Second Brain Publishing.
