> Whether that student walks out the door five years later as an independent thinker or a competent prompt engineer is, institutionally speaking, irrelevant.
I think this is a simplification, of course Bob relied on AI but they also used their own brain to think about the problem. Bob is not reducible to "a competent prompt engineer", if you think that just take any person who prompts unrelated to physics and ask them to do Bob's work.
In fact Bob might have a change to cover more mileage on the higher level of work while Alice does the same on the lower level. Which is better? It depends on how AI will evolve.
The article assumes the alternative to AI-assisted work is careful human work. I am not sure careful human work is all that good, or that it will scale well in the future. Better to rely on AI on top of careful human work.
My objection comes from remembering how senior devs review PRs ... "LGTM" .. it's pure vibes. If you are to seriously review a PR you have to run it, test it, check its edge cases, eval its performance - more work than making the PR itself. The entire history of software is littered with bugs that sailed through review because review is performative most of the time.
Anyone remember the verification crisis in science?
Humans are also unreliable, we are competing for scarce attention, platforms decide what gets visibility and we cater to their algorithms. You could say humans are prompted by feed ranking AI - what and how to publish.
My own approach also has intent sitting at the top: intent justifies plan justifies code justifies tests. And the other way around, tests satisfy code, satisfy plan, satisfy intent. These threads bottom up and top down are validated by judge agents.
I also make individual tasks md files (task.md) which makes them capable of carrying intent, plan, but not just checkbox driven "- [ ]" gates, they get annotated with outcomes, and become a workbook after execution. The same task.md is seen twice by judge agents which run without extra context, the plan judge and the implementation judge.
I ran tests to see which component of my harness contributes the most and it came out that it is the judges. Apparently claude code can solve a task with or without a task file just as well, but the existence of this task file makes plans and work more auditable, and not just for bugs, but for intent follow.
Coming back to user intent, I have a post user message hook that writes user messages to a project scoped chat_log.md file, which means all user messages are preserved (user text << agent text, it is efficient), when we start a new task the chat log is checked to see if intent was properly captured. I also use it to recover context across sessions and remember what we did last.
Once every 10-20 tasks I run a retrospective task that inspects all task.md files since last retro and judges how the harness performs and project goes. This can detect things not apparent in task level work, for example when using multiple tasks to implement a more complex feature, or when a subsystem is touched by multiple tasks. I think reflection is the one place where the harness itself and how we use it can be refined.
claude plugin marketplace add horiacristescu/claude-playbook-plugin
source at https://github.com/horiacristescu/claude-playbook-plugin/tree/main
The hierarchy you describe (intent -> plan -> code -> tests) maps well to how Ossature works. The difference is that your approach builds scaffolding around Claude Code to recover structure that chat naturally loses, whereas Ossature takes chat out of the generation pipeline entirely. Specs are the source of truth before anything is generated, so there's no drift to compensate for, the audit and build plan handle that upfront.
The judge finding is interesting though. Right now verification during build for each task in Ossature is command-based, compile, tests, that kind of thing. A judge checking spec-to-code fidelity rather than (or maybe in addition to?) runtime correctness is worth thinking about.
Yes, judges should not just look for bugs, they should also validate intent follow, but that can only happen when intent was preserved. I chose to save the user messages as a compromise, they are probably 10 or 100x smaller than full session. I think tasks themselves are one step lower than pure user intent. Anyway, if you didn't log user messages you can still recover them from session files if they have not been removed.
One interesting data point - I counted word count in my chat messages vs final code and they came out about 1:1, but in reality a programmer would type 10x the final code during development. From a different perspective I found I created 10x more projects since I relied on Claude and my harness than before. So it looks user intent is 10x more effective than manual coding now.
I built a harness where my plans and code are reviewed with 'claude -p' but most work is interactive, now it has been wrecked. I relied and integrated with Anthropic to get burned. I'm not even maxing out my plan, never surpassed 60%. But now I have to pay API pricing on top? This tells me how trustworthy Anthropic is. If you depend on any specific feature you are at their mercy.
Prior to Anthropic I have had bad experiences with Windsurf and Cursor, same shit - I pay the plan, they shrink my usage quota after a short time, couple of months or weeks. I never returned to Windsurf after they abused me, and never used Cursor after I got my Claude sub, I have no idea where I'll end up next. Too bad Anthropic is pushing my $200/mo away.
> In their minds, financial success is the ultimate yardstick.
In a loopy recursive way, it is. Cost gates what we can do and become. Paying back your costs to extend your runway is the working principle behind biology, economy and technology. I am not saying rich people are always right, just that cost is not so irrelevant to everything else. I personally think cost satisfaction explains multiple levels, from biology up.
Related to introspection - it certainly has a cost for doing it, and a cost for not doing it. Going happy go lucky is not necessarily optimal, experience was expensive to gain, not using it at all is a big loss. Being paralyzed by rumination is also not optimal, we have to act in time, we can't delay and if we do, it comes out differently.
That may or may not be true in aggregate, but for extreme outliers it's impossible to separate from survivorship bias. Are Musk and Andreeson really the most skilled entrepreneurs in the world or are they just good enough for luck to propel them to stratospheric success?
They found luck and success and continue to compound that. However it's easy to make so much money when you have that much already. Just promise the world or invest in companies that do and ride unicorns with private investments into the sunset. The risk they take now is very low.
I feel like they will never suffer the consequences of their actions in any negative way should they get it wrong.
Rarely do we see billionaires not become billionaires because they know how the game is played because they shaped the game so they only ever fail upwards.
> Just promise the world or invest in companies that do and ride unicorns with private investments into the sunset.
Yes, which is why the ranks of the very wealthy are filled with lucky grifters. They got rich by luck, then expanded that wealth with some combination of fanciful statements, lies, and outright fraud.
If you look at the entire entirety of understood history of biology:
The most ruthless always wins
That is to say if I go into a village and kill all the adults and teenagers and steal all the kids who are scared to be killed by me, then I will win in the probably two successive generations that I’ve been able to successfully brainwashing into thinking I’m some kind of God.
That is until somebody kills me and then takes over the structure. For example there are no dictatorships that last past the third generation
That is literally and unambiguously how all life operates
There are intermediary cooperation periods. But if you look at the aggregate time periods including how galaxies form it’s all straight up brute force consumption
That's not how humans came to populate areas that previously were dominated by predators who would be obviously deadly to individual humans. Cooperation and planning are what made physically weak humans dominant. That cooperation and planning developed and flourished without authoritarian structures.
A brief look at certain native American tribes might show quite a lot of talking and consensus building, like if some war chief wants a war he needs to drum up support for that. Hours of talking ensue! Not to say that ancient tribes didn't have the worst of what modern corporations have to offer as far as leadership goes, but a claim "basically every village" is basically wrong, or "bascially" is carrying a heck of a lot of weight.
All Native tribes have been thoroughly dominated and decimated into being constrained to reservations by waves of brutal colonists, that were, as I said, the most ruthless.
Except where the lucky win, or the most ruthless actually suck at winning wars (Bret Devereaux has some interesting observations here, and a study of the various "unbeatable" and of course ruthless empires may also be educational), or where various species cooperate in various ways, or where nobody cries when Mr. Ruthless mutters "rosebud" then dies. "Well, it couldn't have happened to a better chap", said the butler.
Cooperation and Competition are typically temporal periods but only one is extractive irrespective of externalities where cooperation can be mutual but with devastating externalities (ecological collapse)
So unfortunately it’s insufficient to simply be cooperative and the fact that macro level cooperation appears to be rare in the universe
Further, the existing examples of mutual cooperative organizations are so rare as to be non-existent. Humans seem to prefer (or are biologically limited to preferring) competition based social and economic structures.
Read some Charles Mann. Tribal leaders if they can really be described as leaders had to work with consensus and cooperation. Modern society is much more coercive.
Among the Cherokee councils--which included both men and women--unanimous agreement was required for any group decision.
In their society, and in so many others who have been crushed by the forces of empire over the eons, the leaders of the people did not get there by murdering their way to the top. They were respected persons who were elevated to that position by the people.
Tradition tells us the Cherokee did once have a heriditary priestly class, who were called the Ani-kutani, or Nicotani. The people long suffered under their arrogance until one of them went too far, raping a woman while her husband was away. Her husband then amassed an uprising of the people and they killed out the Nicotani to the last man.
(An existence proof that it can be done, if nothing else.)
The primary issue with this is that there is a significant amount of luck involved in acquiring large sums of wealth.
It's hard to get firm numbers around this, but it's estimated around 30-40% of the wealthiest people in the world, derive their wealth almost entirely from inheritance. It's actually very difficult to measure this accurately because a lot of studies will report people as "self-made" even if they started with a small $10 million loan from their parents.
Wealth also follows power laws such that it's significantly easier to acquire more of it once you pass certain thresholds.
Take Mark Cuban - made billions selling some crappy radio service to Yahoo!. Has done effectively nothing since then except for re-investing the proceeds from the buyout. He's technically self-made but it's hard to argue he was anything other than lucky.
Sure, but this argument doesn't actually invalidate the parent at all.
To go back to your biology point:
Figures like Andreessen or Musk (or, at least in my opinion most billoniares) can be directly compared to cancer. They are EXCELLENT at extracting value from the environment they're in. If you limit your moral judgement to just that... then you clearly think cancer is wonderful, since it does the same thing!
Cancer is a group of cells that chemically signal the body to provide resources and spread themselves without restraint, avoiding internal systems that would regulate it via things like apoptosis or other signaling. If you judge a cell by how many resources it can accumulate... Cancer is wildly successful.
But the problem is that extraction without introspection, success with insight, moving without care... eventually actors like this destroy the system they operate within.
Ex - Andreessen should perhaps spend some introspection on the fact that ultimately "dollar bills" are literal cloth (or more likely... digital numbers) that he can't eat, won't shelter him, and can't emotionally satisfy him.
They strictly have value because of the system he operates within that allows exchange, and if he acts without care of that system... he might destroy it. Or it might destroy him.
---
So directly to your point: There is clearly a need for more introspection than "zero". And suggesting otherwise is unbelievably conceited. It is cancerous, and should be treated as such.
I'm damn near broke right now but it would be obvious to you if you spent ten minutes with me that I'm healthier both mentally and physically than either of those two and I can walk down any street with relative impunity and talk with any stranger I meet without concern that they'd recognize me and have beef with me over some stupid shit I did online. I know that when I interact with people it's because they want to interact with me and not my money.
It's true that the cage they live in is gilded but it's still a cage.
Sometimes I stumble across wikipedia biography pages a person like a mumblerapper who had a meteoric rise in fame and wealth only to die in a puddle of puke from a Xanax overdose at like 25. It's sad and everything but when I read it I just think "Man, what a fucking idiot..." Like sure this dude probably had a great few years conspicuously consuming a bunch of shit and showing off a bunch of money with some floozies hanging off his arm but where is he now? Dead and cold in a hole in the ground. And he died a pretty pathetic death to boot.
I don't know about Andressen but I'm pretty sure I'll outlive Musk. As risk adverse as he is for his physical safety he'll end up doing something downright stupid that ends in his untimely death. With Andressen there's a growing possiblity that enough people wise up to his destructive impact on society and a movement where people who are still physically capable but with inoperable brain cancer or something start taking out people like Andressen.
I have yet to check the prediction markets for this proposition but I would bet on Peter Thiel being the first one to mistake a fancy cup for the Holy Grail.
The curse of fame is really underappreciated. Rich and famous people obviously never talk about it in public as it is going against the narrative that builds their brands, but they feel it. They are so jealous of the quietly rich who no one will recognize. Who can still live the same life as you and I. They really are trapped. They basically have to fall of the face of the earth and age out of their appearance to have a chance of obscurity. And their line of work makes that impossible.
They can't go to grocery stores. They can't go to parks. They can't go to casual events. They can't be spontaneous and they can't be serendipitous. Any relationship they have with people is in the shadow of their image. Most people they interact with are trying to grift them in some was as they are a publicly known high value mark. People value what they can get from them vs their personality. Over time they subconsciously under stand this, start to trust no one, and rely heavily on a circle of people who happen to be in reach who may still be grifting them. It is like they live in some artificial habitat on earth, supported by staff, not actually on earth.
Oh come on. So you're rooting for the evil genius in the comic book movie? You would harm millions of people to move up the financial success yardstick?
I don't think many people would agree with such positions.
I do think that people who have succeeded financially might adopt that ethos as an ex post rationalization.
I feel like your comment is evidence that you are insufficiently acquainted with various flavors of cult-like behavior and wingnuttery. There are in fact people who sincerely believe that you don't have to eat [1], who believe it so fervently that they risk and sometimes lose their lives for that belief.
Humans are social creatures. We are biologically inclined to follow charismatic leaders, even off a cliff. In most people, the susceptibility to suggestion is much stronger than the strength of their rational beliefs. Just look at American politics, for example.
All of this is to say that if Andreessen said, "I don't eat food," there would be a small but vocal group who would see that as validation of their beliefs; there would be a think-piece in the Atlantic about the history of breatharianism; Hacker News comments about what does "food" mean, really, etc. Yes, people would take it seriously. Just because he's rich and has therefore bought a loud megaphone.
Are you kidding? People would eat that up if he said that. Soylent would sell like crazy. You'd see protein smoothie shops pop up all over the bay area. For better or worse there is a subset of people who just lap up at whatever comes out of these people's mouth.
If you were to make a list of the most important people in history how different would it be from the list of the richest people in history?
How many different occupations do you think you would find on the important list. Would it have scientists, mathematicians, doctors, engineers, world leaders, activists, religious figures, teachers?
How many occupations do you think would be on the richest list?
Do you think it is fair to judge the success of Martin Luther King Jr or Albert Einstein based on the amount of money they made?
I tried that for a couple weeks and it's no where near as well integrated as Cursor. I hope they get there though because I like Zed.
Zed plus Claude feels more like using isolated browser extensions instead of something part of the browser (unless you pay for Zeds AI thing then the integration is marginally better).
The cognitive benefit of writing comes from externalizing and evaluating ideas under friction. LLM conversation provides more friction per unit time than solo drafting because you're constantly reacting to a semi-competent interlocutor who gets it almost-right in ways that force you to articulate exactly where it went wrong.
I checked my logs and I write 10 words in chat for 1 word in LLM output for final text. So it's clearly not making me type less. I used to type about 10K words per month now I type 50-100K words per month (LLM chat is the difference).
The surplus capacity provided by LLMs got reinvested immediately in scope and depth expansion. I did not get to spend 10x less time writing.
> it certainly feels against the spirit of what I intended when distributing my works
You can own the works, but not the vibes. If everyone owned the vibes we would all be infringing others. In my view abstractions should not be protected by copyright, only expression, currently the abstraction-filtration-comparison standard (AFC) protects abstractions too, non-literal infringement is a thing.
Trying to own the vibes is like trying to own the functionality itself, no matter the distinct implementation details, and this is closer to patents than copyrights. But patents get researched for prior art and have limited duration, copyright is automatic and almost infinite duration.
I think this is a simplification, of course Bob relied on AI but they also used their own brain to think about the problem. Bob is not reducible to "a competent prompt engineer", if you think that just take any person who prompts unrelated to physics and ask them to do Bob's work.
In fact Bob might have a change to cover more mileage on the higher level of work while Alice does the same on the lower level. Which is better? It depends on how AI will evolve.
The article assumes the alternative to AI-assisted work is careful human work. I am not sure careful human work is all that good, or that it will scale well in the future. Better to rely on AI on top of careful human work.
My objection comes from remembering how senior devs review PRs ... "LGTM" .. it's pure vibes. If you are to seriously review a PR you have to run it, test it, check its edge cases, eval its performance - more work than making the PR itself. The entire history of software is littered with bugs that sailed through review because review is performative most of the time.
Anyone remember the verification crisis in science?
reply