[Article] “Working in US AI policy” by Niel Bowerman

The US Government is likely to be a key actor in how advanced AI is developed and used in society. Learn more about this career path.

Narrated by AI.

---

Outline:

(01:12) Summary

(02:56) The aims of AI policy

(04:11) Advanced AI as a coordination problem

(05:22) Definitions: artificial general intelligence and other types of advanced AI system

(06:14) Key actors

(08:02) Which parts of AI policy do we think are most important to work on?

(10:41) Policy careers involve moving between different types of roles and institutions

(12:13) Why pursue this path?

(13:33) 1. The US government is a powerful actor that may play a key role in the development and deployment of advanced AI systems

(13:43) The US government is likely to be involved with the development and deployment of advanced AI systems

(13:50) AI and society

(15:09) AI and national security

(16:21) The US government is a powerful actor

(16:25) The US government is well resourced

(17:21) The US government has unique abilities

(18:08) 2. Individuals can have a large positive impact in government

(18:48) Can individual federal employees influence the government?

(20:36) How large are the budgets that government officials oversee?

(22:37) What are your chances of moving into senior roles in government?

(24:03) Comparing the competitiveness of government roles against those in industry

(25:08) How meritocratic is career progression within government?

(25:43) 3. There is an AI skills shortage in the US government

(25:50) AI skills are in demand in government

(27:04) Not enough people in government are thinking about catastrophic risks from advanced AI systems

(28:01) There are many more roles available for talented people

(29:24) 4. Government careers were already a good default option

(31:07) Arguments against

(31:11) Your social impact in this career path is high risk, high reward

(32:51) You need to be in the right place at the right time

(33:49) Governments can be difficult to predict

(34:32) International coordination problems often haven't been solved historically

(37:21) Will you be able to influence any critical junctures associated with the development of AI?

(38:36) It is probably better to initially work in AI policy at a top AI lab if you can

(40:05) Other arguments against

(41:34) How to pursue a career in US AI public policy

(41:54) Where to aim long-term

(41:58) Political appointees and the Senior Executive Service

(43:28) Which parts of government are most relevant?

(43:44) US Congress

(45:28) US executive branch

(48:47) Fellowships

(49:19) Other organisations

(50:09) Key routes in

(54:16) Find opportunities on our job board

(54:24) Be careful to avoid doing harm

(56:19) Fit: Who is especially well placed to pursue this career?

(56:51) What predicts fit?

(57:02) Have relevant expertise in an area relevant to AI policy

(58:01) The importance of relationships

(58:59) Conventional credentials matter

(01:00:13) Governments are not nimble

(01:01:07) Good judgement and calibration

(01:01:36) Other traits

(01:03:24) Other personal fit considerations

(01:04:40) Which paths could be even better?

(01:06:03) Conclusion

(01:07:26) American interested in working on AI policy?

(01:07:51) Learn more

(01:07:55) Getting started in this career

(01:08:38) More about AI policy

The original text contained 33 footnotes which were omitted from this narration.

---

First published:
January 31st, 2019

Last updated:
July 31st, 2023

Source:
https://80000hours.org/articles/us-ai-policy

---

Narrated by TYPE III AUDIO.