Summary: If we validate automated alignment research through empirical testing, the safety assurance work will still need to be done by humans, and will be similar to that needed for human-written alignment algorithms.

Three levels of automated AI safety

Automating AI safety means developing some algorithm which takes in data and outputs safe, highly-capable AI systems. Let's imagine three ways of developing this algorithm:

1. Human-written algorithm, AI details: Humans write down an overall AI safety algorithm, and use AI systems to fill in a bunch of the details. The humans are confident the details provided by the AI systems don’t compromise the safety of the algorithm. This category includes scalable oversight, semi-automated interpretability (LLMs explain each neuron or SAE feature), and using LLMs for scaled formalisation of a spec.
2. AI-written algorithm, checked empirically: The humans might have some rough idea what overall scheme is good, but the AI is going to [...]

---

Outline:

(00:19) Three levels of automated AI safety

(03:09) A model of empirically checked automated safety

(07:18) We should understand if the collapsed alignment scheme is safe

(09:45) We should ask if less automation is possible

(12:11) Checking the algorithms avoids collapse, but is hard

(14:11) Acknowledgements

The original text contained 1 image which was described by AI.

---

First published:

Source:

Narrated by TYPE III AUDIO.

---

Images from the article:

Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.