Previously: On RSPs.

Be Prepared

OpenAI introduces their preparedness framework for safety in frontier models.

A summary of the biggest takeaways, which I will repeat at the end:

I am very happy the preparedness framework exists at all.
I am very happy it is beta and open to revision.
It's very vague and needs fleshing out in several places.
The framework exceeded expectations, with many great features. I updated positively.
I am happy we can talk price, while noting our prices are often still far apart.
Critical thresholds seem too high, if you get this wrong all could be lost. The High threshold for autonomy also seems too high.
The framework relies upon honoring its spirit and not gaming the metrics.
There is still a long way to go. But that is to be expected.

[...]

---

Outline:

(00:07) Be Prepared

(02:48) Basic Principles

(07:33) Veto Power

(10:27) Introductory Section and Risk Categories

(13:13) Cybersecurity

(15:58) CBRN (Chemical, Biological, Radiological and Nuclear) Threats

(18:47) Persuasion

(22:24) Model Autonomy

(25:34) Key Takeaways From Risk Descriptions

(28:36) Scorecards

(31:27) Governance

(34:56) Deployment Restrictions

(36:21) Development Restrictions

(39:50) Conclusion and Biggest Takeaways

---

First published:
December 21st, 2023

Source:
https://www.lesswrong.com/posts/hQPfLsDKWtdvMwyyr/on-openai-s-preparedness-framework

---

“On OpenAI’s Preparedness Framework” by Zvi

LessWrong (30+ Karma)