AI Safety Fundamentals: Governance

12 Tentative Ideas for Us AI Policy

May 13, 2023

A discussion on 12 US AI policy ideas to improve outcomes, covering topics such as governance, advanced AI regulation, harm tracking, and emergency shutdown mechanisms.

09:49

AI Summary

AI Chapters

Episode notes

Podcast summary created with Snipd AI

Quick takeaways

Control export of Frontier AI models to limit risks.

Enforce strict information security and safety testing for AI models.

Deep dives

Control the export of Frontier AI models

One key policy recommendation is to control the export of Frontier AI models that possess highly general capabilities. This involves restricting the export of models trained with a significant compute budget, contributing to limiting the spread of models posing high risks. Moreover, regulating API access can prevent the generation of optimized datasets for training risky models efficiently, enhancing governance over AI proliferation.

Introduction

4min

Policy Recommendations for Advanced AI Governance

4min

Policy Suggestions for AI Harm Tracking and Shutdown Mechanisms

2min

About two years ago, I wrote that “it’s difficult to know which ‘intermediate goals’ [e.g. policy goals] we could pursue that, if achieved, would clearly increase the odds of eventual good outcomes from transformative AI.” Much has changed since then, and in this post I give an update on 12 ideas for US policy goals[1]Many […]

The original text contained 7 footnotes which were omitted from this narration.

---

Source:
https://www.openphilanthropy.org/research/12-tentative-ideas-for-us-ai-policy---

Narrated by TYPE III AUDIO.

A podcast by BlueDot Impact.

Learn more on the AI Safety Fundamentals website.