Ability to see expertise is a milestone worth aiming for (2022)

I sat next to an Econ professor (at an Ivy) on a flight. I told him I had worked at a hedge fund, and he was like "wow, I could never put myself in a situation where my living depended on my economic theories being right."

A good friend of mine is a kinda-famous evolutionary biologist. He's in his mid-40s now, unmarried and doesn't have kids, and likely none on the horizon.

In the economist case, what does it mean that in his expert opinion, he wouldn't trust his own expert opinion! In the biologist case, what does it mean to be one of the world's leading experts on "evolutionary biology" and yet completely fail to pass your own genes?

What's most interesting to me is that neither of these scenarios strike us as "super wrong." I can imagine most people wouldn't hesitate to attend a class taught by either one of these guys, while we'd avoid analogous situations (eg: we wouldn't use a broke financial advisor or fly on an airplane whose designer refuses to set foot on it.) What's the difference?

This to me sounds like a simple example of he understands the limitations of his work!

As for your evolutionary biologist friend. Being interested in evolutionary biology does not mean you are inherently someone that wants to be a parent. These are very very different things. Besides, evolutionary biology impacts more than just humans, it impacts every species. Being a parent comes with a great responsibility. Someone could just as easily be a really big fan of dogs, but not want one because of the responsibility or the costs, or similar.

Because there are two parts in expertise: pure knowledge, and a set of weighed criteria to evaluate things. This second part is what I think TFA calls "tacit" knowledge. This stuff is a job for neural networks, but more for Alpha0 than for ChatGPT - although it does probably that under the hood, but it's an expertise used to string together questions and answers. That's enough to give the illusion of expertise, but you still face the problem how evaluating that artificial expert. There are many reports that ChatGPT can be "slightly wrong" that only experts can really spot. You can have ChatGPT do your boilerplate code or bulk work if you are an expert in the field.

I personally think of this part of expertise as the expert's "intuition". This intuition is indeed sometimes difficult to communicate. Sometimes because it comes from various experiences, and describing those experiences in enough details would take far too long. Sometimes A just "feels" better than B because those criteria and weights are subconscious; pretty much like you apply a certain amount of force to lift an object because you have subconsciously evaluated its weight (either with material density and shape, or from memory because you've lifted similar things before) - and sometimes you are surprised because it is magnetized or it the box has been emptied.

However, my criteria to evaluate expertise is still how well the expert can explain something to me - that is, if they are good teachers. This also an unreliable proxy like the "vocab point" of TFA. Not everyone is a good teacher, and impostors or poseurs can make you feel like you understand when they are actually bullshitting you.

ChatGPT could be a good (enough) teacher that can help you becoming an expert (partly because of the fact checking work you'll have to do, something you normally don't have to do with an actual teacher), but ChatGPT as an expert, it is a no-no in my book.

I can, and have, explained 'deeply technical things' in 'business terms'. And people understood. Then made a bad/poor/wrong choice anyway, because... "well... all those things you're saying might happen, they might not. Or... it may not happen for months or years, and we already committed to date X for project Y".

I can tell people the cost of punting on 'deeply technical thing' will be enormous, like... "this will require doubling the dept headcount and stopping all new progress for 6-12 months". Then offer alternatives. Then get summarily ignored, and watch things crash months later, and the dept is then stuck for months scrambling to understand all the problems they could have avoided. The decision maker who decided for extremely short term gain? They're gone (sometimes of their own choosing, sometimes not).

The "risks" can seem wildly out of touch with reality, but many times if you're making poor choices you do hit the "worst case scenario" wall. Colleague has been contracting inside a shop on and off for a few years. He indicated "if we don't do X, we will not be able to do any new feature development for at least 18 months". And... closing in on 4 years later, they've not shipped any substantial new features (some bug fixes, but that's about it).

I've seen this happen once up close, and have seen inside other companies and got the war stories from others.

Does this happen every single time? Of course not. But even if you can confidently and correctly and pleasingly describe deep tech things to business folks in their own language, showing the pros/cons/costs/risks... that doesn't mean they will take your side. Saving face, hitting deadlines at all costs, and other factors weight on decisions, and you can't control that. You can control where and how you work though.