You Ain't Never Had a Friend Like GPT

The need for AI mentorship

Brine Test Jug comments,

I want an AI mentor that I can walk with through the woods. He already has a good idea of who I am because he’s read my entire internet history. He’ll strike up a conversation and learn everything else about me as we walk. He’ll learn that I’m not very good at math proofs, and I’ve never had the patience for them. During our walk we decide that we’ll work on proofs. He tells me exactly what I need to hear, exactly when I need to hear it. He senses that I’m getting frustrated, and immediately changes to the next most likely method to get me to understand. I never need to sit down at a chair. I never get frustrated while learning. My attention span is as maximized as any AI fueled TikTok binge. AI can be the mentor that teaches anybody anything, regardless of their ability to sit down and pay attention.

Since I wrote about AI winners and losers last month, I’ve spent a lot of time trying to sell friends and family members on the idea that ChatGPT is going to radically change access to mentorship. But so far only BT Jug seems fully on board with me.

I’ll lay out my case over two posts, starting with this one on why the pre-GPT status quo is so awful.

II. Mentorship is WAY under-supplied

Mentorship creates a lot of value for the world. But there are good reasons why most people are reluctant to provide very much of it.

Put yourself in the mentor’s shoes—

1. You have so many other things on your plate. Time spent mentoring is time that you can’t spend scoring the next promotion, racking up billable hours, or writing your memoirs.

2. There’s limited upside. The protégé captures the bulk of the benefit if things really take off, and may even end up competing with you!¹

“12 Successful People Who Started as Interns”

3. Respect, admiration, prestige? Unlikely except in some very narrow ways. Most people are like George Costanza.

4. No good deed goes unpunished. The more willing you are to help someone the more likely they are to make additional requests. You already receive too many emails.

5. Often things do not work out how you hoped. And the more invested you are in a person, the more it will hurt if they fail. On the other hand, it’s hard to feel guilty for the success someone didn’t have…

When you look at it this way, it’s a wonder that mentorship happens at all.

III. Exceptions that prove the rule

Some might be inclined to dismiss my argument because it proves too much. After all, most people reading this likely have some experience with being liberally mentored.

For instance, we do see a lot of mentoring between…

• Mothers and their offspring

• BYU professors and high-potential undergrads

“BYU is Top-5 Launching Pad for PhDs”

• GMU econ bloggers and their fans

Econlib

But in each case, the principles from section II do a lot to explain why so much mentorship takes place in these exceptional cases. Consider that…

1. For biological as well as cultural reasons, women often compete for status indirectly.

Alice Evans @_alice_evans

Women’s proclivity to compete depends on the prize. If their kids stand to gain, women get competitive. pnas.org/doi/10.1073/pn… Evidence from 🇨🇳, by @Cassar_Ale, Wordofa, & Zhang

2. BYU undergrads are much more likely to return to their alma mater²

3. GMU is often shut out from traditional forms of academic prestige

“The Ideologues of GMU”

The upshot? If you think mentorship is awesome in these exceptional cases, you should be even more convinced that mentorship is under-provided in a lot of places.

IV. The Demand-Side

Now put yourself in the position of the would-be protégé. For many such people, almost all of the time, things pretty much suck.

Often mentors are not to be found at all, or are of dubious quality. Look at how many young people find themselves turning to social media influencers for guidance.³

Richard V. Reeves @RichardvReeves

"If we don’t like Andrew Tate’s script, we need to provide a better one, rather than simply railing against theirs." Me on the West's lost boys for @unherd

unherd.comAndrew Tate and the West’s lost boysYoung men are struggling in a woman’s world

Even those lucky enough to gain access to a decent mentor must usually spend a lot of time doing obscenely boring and low-value tasks before they can get any real investment in return.

How high is transfer of learning here realistically?

And of course it is exceedingly rare to find someone who fully believes in you when you’re an unusual person or a risky bet along some dimension. If it weren’t for Dr. Dre, Eminem might never have been discovered.

Link. See also 50 Cent—“I had to run on my own energy” for six years.

V. GPT Mentorship in Perspective

When friends or family members object to my vision for AI mentorship, the comparison they have in mind is usually an excellent real-life mentor. But such a blessed alternative is seldom the most relevant point of comparison.

Since any request I make of a human mentor in some sense draws down on their willingness to help me with something else, I often find myself wasting a lot of time reducing the cognitive and emotional burden of my request with careful editing or else going about things the hard way knowing that a single informed remark likely could have solved the entire dilemma.

But AI never leaves my emails unread or looks down on me for asking “a dumb question.” Just while working on this post, AI helped me improve my writing and have time to write in the first place.⁴

Data redacted.

And though such acts of mentorship may seem small and simple, to me they are already worth appreciating. To get where I’m going, I’ll need all the help I can get.

---

Thanks for reading! If you enjoyed this, you can subscribe for part 2 or share this article with a friend.

Your comments are also appreciated, and have a good chance of influencing my next post about who stands to benefit the most from GPT mentorship and how to recognize such opportunities in your own life. I am very grateful to BT Jug for inspiration on this one :)

1. Steve Jobs aside, almost anyone you invest in becomes somewhat more likely to demand a raise or move on to greener pastures. This is known as the hold-up problem.

2. See for instance their economics department.

3. To be clear, there are also many good ones!

4. Again, super simple, but these little things really make me more confident.

Infovores @ageofinfovores

I've always been just a little unclear about what "an exception that proves the rule" means, but GPT put this so clearly for me.