Nobody will buy an AI with enough context to develop critiques of their own organizational structure.

I've been professionally trained to monitor my own thought process and review my notes for signs of bias, and I've spent decades absorbing new domains well enough to build testable models. When I look at understanding political issues the people I rely on to help me "put forth the work" are gone, man. The effort I need to put in on one subject well enough to make decisions now is immense.

If you don’t need the machine to respond instantly (or explain your own business model to you) everything can be local and it’s been like that for a few years now.

There’s a reason that most of the voters (and protesters in my area) are retired, and it isn’t apathy. I don’t have time to educate myself on these topics in any real depth.

And I need to educate myself because the push information is all bullshit. Digging into policing in Seattle, the official and public conversation was all culture war while the actual problems looked like simple incompetence from a system analysis perspective.

I don’t have the bandwidth to deal with this kind of fumbling on every topic, and I’m realizing that my parents didn’t live in a low-trust society like I do.

The problem with your point of view is that "Love of cocaine" is one of them, it's near the top, and you'll never acknowledge the fact.

Our documented examples of voter fraud come from a time when in-person voting was the only option, again something we teach in school, while the modern concerns from security professionals focus almost entirely on electronic voting machines.

Interesting to think about the cost of training a LLM to understand that it’s operating within an unknown number of larger contexts versus sending that quote to an edgy intern.

"Involve transgenics" is broad enough cover the intern doing literature reviews on related subjects. The discovery process on an unfamiliar domain is a jargon/term of art minefield, and the phrases that fly past me turn in to shibboleths.

Impossible as you said. Context isn’t static, it’s continuous, analog, and a conglomeration of viewpoints.

AI cannot create useful context for itself because it is a machine with no desires. It doesn’t have a point of view, it has historical records. It moves forward in time by walking backwards (if that makes sense?)

Because of this there's a chain of trust between myself and the tools I rely on to do work. The people who create those tools see unpredictability as a problem, and that's the only reason I'm using them. I can't work on important systems with a vendor product like Claude Fable.

That being said there's plenty of work to do where it'd be amazing. This isn't an either/or situation.

What kind of Northern Exposure bullshit is this?

The fancy plan.

Let me introduce you to the phrase "I don't see a mechanism."

If I’m using a calculator to verify my math, I don’t want to use a second calculator to verify the first one.

As far as I can tell this part of the job isn't really on anyone's radar anymore.