This is a timed post. Every 5 minutes while writing this post, I had to stop to do 15 push-ups, and when I could no longer complete my required number, I had to upload it. My friends thought this would be a fun challenge, but it means that it will likely be less polished than some of my other output.
Summary: There has been a lot of talk in the EA/AI safety community recently about what it means to be a “generalist,” but many disagree on what this means. In this post, I give my take on the role: a generalist is best understood as someone who does whatever needs to be done to solve a set of problems, rather than as a specific career role.
What is a generalist?
There’s been a lotoftalk lately about the need for “generalists” in AI safety. However, people across the space seem to have different understandings of what it means to be a generalist. Some seem to believe that “generalist” is just a weird title to describe a lump of roles that don’t fit anywhere else. Others seem to think it’s a synonym for “fieldbuilder” or “impact multiplier.”
While both of these definitions gesture at what a generalist is, I think it’s wrong to conceive of a generalist as a specific set of tasks. Instead, I think that a generalist is someone who looks for the most important problems to solve and shapes themself across a broad range of skills to address those problems.
I don’t think of this as a definitional shift for what it means to be a generalist. Instead, I view “being a generalist” as a disposition to seeking out the most impactful thing across a broad range of skills, similar to how “being an optimist” is a disposition to thinking good things will happen in the future. This might mean that you can’t easily picture a pig wearing clothes in a children’s book being a generalist, but I think that’s okay.
Success as a generalist requires agency. Ben Kuhn has my favorite explanation of what this means[1]: agency is working backwards from your end goal and doing everything necessary to achieve it without waiting for someone to give you permission. I like this explanation for a few reasons:
If you don’t understand the problems in the world and why they exist, it is unlikely that your work will address them.
Many problems aren’t being solved because either nobody has noticed them or because nobody else has time to address them. Unless someone steps up to fix them, most of the time, they will not be solved.
Here are some examples of people being generalists:
I’m contracting as Operations Lead for BlueDot’s Incubator Week. As a result, I spend a lot of time thinking, “What makes Incubator Week impactful, and what can I do to increase that impact?” This has led me to seek additional feedback from past participants and pre-empt future problems by providing programming that addresses these issues.
One of the Generator residents noticed that many people in AI safety were struggling to find the job roles that fit them best, so they made it their mission to match everyone with the roles they were best suited for. This resulted in several valuable connections!
Due to unrelated circumstances, I’ve come across several people interested in AI safety at schools without AI safety university groups. So, I’m helping them start groups. If I hadn’t been thinking about broad field-building, I don’t think I would have done this.
A group of former debaters was trying to bring more talent into the field when they realized that other top debaters might be sympathetic to the ideas and great strategists. From this, Praxis was born, and its results are looking promising.
One generalist noticed that much of the discourse informing policy outcomes was chronically online: exchanges about how AI should be regulated were happening over Twitter threads, where nobody could see eye-to-eye. To address this, she brought everyone together in person, where they could discuss and relate to each other more easily.
I know many generalists who have found promising young people who weren’t getting the proper mentorship required to grow and improve, and have taken it upon themselves to address these needs.
AI 2027 and Plan A were born when a group of generalists realized that the public lacked awareness of the threats the upcoming years of AI could hold and how to address them. Then, some of the other generalists working with that team looked for additional opportunities to make sure it was seen by as many people as possible.
I think that almost every outcome from the Generator Residency this summer has been really awesome.
My hope is that you, as a reader, come away from this post with the intention to address real problems you find in the world.
Some quick nuances and clarifications:
When Kairos says that we need generalists, they’re pointing at a real need in AI safety for people who fill a certain set of job roles. When they use the word “generalist,” they’re pointing toward those job roles, and when I use the word “generalist,” I’m pointing toward an approach to doing impactful things, including those job roles.
Sometimes people aren’t doing certain things for good reasons, so you should think carefully about what you’re doing before sinking too much time into them.[2] Also, many things aren’t very good ideas. If you’re just starting out as a generalist, you’ll probably lose a bunch of time pursuing things that aren’t that impactful, and that’s just part of the process.
“Can’t anyone be a generalist by this definition, even if they’re a technical researcher?” Yes. I’m using this post to describe how I would like people to think about doing impactful things.
Push-ups
Upon completing the first draft of this post, I had done 405 push-ups
I gave up after 435.This is the most I’ve struggled with the push-up part of one of these posts.
Cate Hall’s definition of agency is something closer to “noticing all the degrees of freedom of life, and then acting on them.” I think Hall gives great life advice, but Kuhn’s explanation is more helpful for someone trying to achieve a specific goal.
This is a timed post. Every 5 minutes while writing this post, I had to stop to do 15 push-ups, and when I could no longer complete my required number, I had to upload it. My friends thought this would be a fun challenge, but it means that it will likely be less polished than some of my other output.
Summary: There has been a lot of talk in the EA/AI safety community recently about what it means to be a “generalist,” but many disagree on what this means. In this post, I give my take on the role: a generalist is best understood as someone who does whatever needs to be done to solve a set of problems, rather than as a specific career role.
What is a generalist?
There’s been a lot of talk lately about the need for “generalists” in AI safety. However, people across the space seem to have different understandings of what it means to be a generalist. Some seem to believe that “generalist” is just a weird title to describe a lump of roles that don’t fit anywhere else. Others seem to think it’s a synonym for “fieldbuilder” or “impact multiplier.”
While both of these definitions gesture at what a generalist is, I think it’s wrong to conceive of a generalist as a specific set of tasks. Instead, I think that a generalist is someone who looks for the most important problems to solve and shapes themself across a broad range of skills to address those problems.
I don’t think of this as a definitional shift for what it means to be a generalist. Instead, I view “being a generalist” as a disposition to seeking out the most impactful thing across a broad range of skills, similar to how “being an optimist” is a disposition to thinking good things will happen in the future. This might mean that you can’t easily picture a pig wearing clothes in a children’s book being a generalist, but I think that’s okay.
Success as a generalist requires agency. Ben Kuhn has my favorite explanation of what this means[1]: agency is working backwards from your end goal and doing everything necessary to achieve it without waiting for someone to give you permission. I like this explanation for a few reasons:
Here are some examples of people being generalists:
My hope is that you, as a reader, come away from this post with the intention to address real problems you find in the world.
Some quick nuances and clarifications:
Push-ups
Cate Hall’s definition of agency is something closer to “noticing all the degrees of freedom of life, and then acting on them.” I think Hall gives great life advice, but Kuhn’s explanation is more helpful for someone trying to achieve a specific goal.
But don’t waste too much time thinking.