To be fair I think its far easier to automate leadership jobs despite how they believe in greater self-worth. As a developer I keep running into problems that require some measure of creativity to solve or workaround. And regardless of how many years I've been in the industry I'm not running out of issues. They just keep getting more obscure. On the other hand leadership communication and problem resolution seem like the exact opposite of creativity and innovation. Its all about time proven, reliable solutions. Same layoff notices, recommendations, design everywhere. Repetitive, boring, politically correct and therefore easy to automate. Accountability will likely become reason n1 for keeping leadership roles intact that is until people become accustomed to AIs everywhere. After all what's the difference between a bot showing a layoff notice at 5am or an AI generated message/video from a human doing the very same thing.
The value-add for a meatbag CEO is the physical networking they do. They have to do executive functions as well, which could quite possibly be replaced with a well-trained LLM and even possibly exceed in overall quality. But an LLM can't go to conferences; they can't have Top Secret Meetings where a merger or acquisition is discussed; they can't run into Jim Executive of BigCorp Inc. in Vail and scheme up some big new thing; they can't fill a room full of institutional investors with confidence that the future is looking quite rosy.
Now, one can certainly argue that all of that is bad, or not even necessary, but it is a thing they do.
But I still think this is a good project, especially if it is demonstrated to excel at some significant fraction of what a meatbag CEO does. Because meatbag CEOs need to know that they are just as vulnerable to this sort of thing as their employees they so callously discard.
Being facetious of course, but didn't AI create their own reddit/chat rooms, what's stopping AI CEOs from "running into" other AI CEOs in other places?
The system works because it runs on human ambition. If people in power handed their control to an AI there wouldn't be companies and mergers anymore. It wouldn't make sense. That world could be great or it could be dystopian.
I would argue the world is already dystopian and that human ambition has more or less doomed the species to extinction. So anything different seems great to me! :)
I am not sure if that is the main value add. I have repeatedly heard that taking responsibility and being on the hook for the decisions (even if AI hands you the decision and all the research and homework that went into it) is the most important feature.
I guess starting with and/or developing a good network is one of the main ways they can gain the confidence for taking responsibility.
I have yet to see a CEO actually take responsibility for anything. Its always some hand-wavey talk about strategy and unexpected decline in sales. Never once was it "I messed up."
I have, on the other hand, seen boards pile all the blame onto the C-Suite. Seems easy enough to pile the blame on an AI.
> they can't have Top Secret Meetings where a merger or acquisition is discussed
This actually seems completely doable for a properly configured LLM. There will probably be a platform in the future where companies can setup LLM to discuss business deals with other companies' LLMs and bring the resulting proposals to the human leadership. Every morning the CEO gets an email about the 3 most promising potential deals that were discussed, a summary of the big picture with TBD details highlighted, and a link to the transcript of the conversation. Plausibly the potential deal is with a company the CEO had never heard about or dealt with.
> To be fair I think its far easier to automate leadership jobs despite how they believe in greater self-worth.
Speaking as someone who has worked at multiple companies all the way from IC to team lead and even up to senior manager with a global team:
Almost every job at every level has some version of "ok, what's the flowchart for this". People assume this is for tech jobs but they would be surprised that it's true for managerial roles as well.
e.g. you have a poor performer then you run through this checklist:
1. is it an issue with the work being given?
2. is it an issue with another co-worker?
3. are they having a family issue?
4. it's a family issue, see if you can get details
5. oh, it's a sick parent? then implement the sick parent policy e.g. more flexible work hours, reduced work load, check in with them weekly till things improve etc
Where the human part comes in is where you have some combination of the below:
- event no one at the company has seen before
- the above requires discussions amongst humans to figure out
- those humans have to talk to other humans outside the company
examples at two levels might be (both for hurricane Sandy):
a. SREs have to deal with datacenters that don't have diesel
b. HR has to deal with critical employees that have houses that were swept away
I guess an AI could propose solutions to the above but it's a human that would actually make them happen (plus humans are more resilient to power outages).
The way you have expressed it there are optimized workflows not only the managerial/executive, administrative, content producing set of workers, but in almost every profession, encoded in beaurocracy, checklist, best practices and culture: doctors, care workers, lawyers, policing ...
The question is how well language models have internalized the optimal decision pathway and mitigating strategies to deviating circumstances. What data do they have to be trained on? If an AI defeats the best Go players in the world, surely it is a good question whether they can exceed in games with large unknowns such as business administration.
As mentioned somewhere else in this thread, the moral and ethical layers is where some of the challenges of these system can be found.
That ends up going in the DeepMind philopsophy of making AI, rather than the OpenAI philosophy. DeepMind makes AI that can defeat the best human Go players, OpenAI makes LLMs that are much more versatile but struggle with finishing a chess match without making illegal moves, never mind winning.
The other question on my mind: Cases where we have known decision pathways is exactly what the expert systems of the 70s and 80s were about. And from my understanding there were some successes, but we mostly found that it doesn't scale. Both encoding optimal decisions and encoding the decision space of things that can happen are huge challenges. LLMs might be genuinely better here, because they can just make things up as they go along, but they are also difficult to control
> I think its far easier to automate leadership jobs
I have the opposite opinion. I've never been a CEO, and have intentionally avoided any sort of manager/C-level opportunities at all the various companies I've worked specifically because they require a lot of effort at personnel management, investor relationships, customer relationships, and generally things requiring having to talk to people.
That kind of thing just can't be replaced by a chat bot.
Having been an engineer and then a founder (and CEO), neither can be automated. Both require organizational work to be effective, and developer productivity is so profitable in an outsized way that you'd always just hire more if you could, even if they were all 10x as productive (maybe this changes at some scale).
I'd have loved for an AI to assist with a few things as a CEO but we would have probably just shipped faster and it would have created even more pressure for board communication, customer calls, design partnerships, legal discussions, compliance, etc.
The idea that AI is coming after leadership is truly laughable. The idea that it's coming after some engineering jobs is a bit more likely to me, but I think many will actually be more valuable.
> Having been an engineer and then a founder (and CEO), neither can be automated
Can it be fully hands-off? No. But I think, as will all things, their both subject to the rise of the "technician"s. A mechanic doesn't change your brakes or flush your oil, a technician does. You don't see a doctor, you see a nurse practitioner and an MD just signs off. All the work that used to be done by someone with schooling to learn how to do it is just done by a technician that learned how to use the tools without the critical thinking of the expert, but with an expert somewhere available if needed.
There's a scene in Breaking Bad, where after having shadowed Walter White as he ran a cook - one of the henchmen makes a batch insisting that it's just following a recipe (to which Walt describes all the things that could go wrong if the operator didn't understand the nuances).
American's can't help but try to automate our jobs away. We create systems, recipes, toolkits etc trying to minimize the amount of actual thinking required to do anything. Honestly, I think that's one of the reasons software quality has gone down across the board. We're already so many abstractions removed from doing the actual work, that most developers don't even know how to debug the systems beyond the highest layer. I've seen the rise of bootstrap devs, and jQuery devs, and react devs, and now it's vibe coders and harness engineers or whatever the latest things are. My examples are dev centric because that's my chosen role, but just look to say Amazon and their use of AI for performance evals and terminations. It effects management less technical jobs too.
> developer productivity is so profitable in an outsized way that you'd always just hire more if you could, even if they were all 10x as productive (maybe this changes at some scale).
Definitely. Software is a digital arms race between companies so they’ll just keep expanding. I’m sure at some scale that breaks down but we are far from that point. My lumberyard, a few miles away from one of ASML’s US facilities, still runs its POS and inventory tracking on old DOS machines. In a decade they could probably make more money selling them as antiques.
A lot of the value (compensation cost) of having a single leader is that the leader becomes a critical, irreplacable bottleneck and silo of information. AI can solve that by removing the bottleneck.
Exactly. Being a good CEO means reading the body language of your programmers so you can tell when they need support. It means communicating competence to investors and customers. It means making complex decisions with incomplete input, a lot of the time deep intuitions are required. It means having the good sense of judgement.
Part of the time you are tyrant, part of the time you are parent, part of the time you are psychologist, mentor, master, healer, empath, judge and jury, and the most important part : is knowing that ALL of this is REQUIRED to be human actions and a little text input spitting out (even under ideal conditions - excellent) quality text to the above is definitely not going to work because every good CEO knows that its 50% the message you deliver, and 50% the delivery that matters.
And also being a technical founder with 20 years coding experience, I dont think AI can replace programmers for a set of different but equally valid reasons.
I know this is mostly true at huge companies, but having spent my career at various "small" corporations, with maybe a few thousand employees total (and only ~50 of which were in IT). I was always surprised when the CEO knew names of people like me, the lowly developer 5 layers below them on the org chart, but beyond that, remembered the last conversation we had, or an interest we might have shared (golf, hiking, travel, etc.).
Even as a temporary consultant, at a company for only a couple of months, I remember seeing the CEO walk the hallway, greeting every single person on the way between a meeting room and the bathroom. I can barely remember my family member's names, or the last thing we talked about, let alone have the memory space for all that information to seem kind and courteous to "low level" employees.
In my entire career, I've never worked at a behemoth-sized company, though, so my experience has been much more "intimate" environments, and primarily in consulting firms or satellite offices away from the main HQ (even by only a couple blocks some times). So everyone in the office knew each other.
I worked at a 10,000 person company ~15y ago. I was a lowly employee at the time; just a run of the mill Research Analyst writing code in SAS. I got on the elevator w/the CEO, said hello, and he knew exactly who I was and what I did.
Two important grounding items for me when I went into management and my career accelerated:
1. Take the worst boss I ever had (coincidentally at the same company) and make sure that I do the exact opposite of everything they did when leading others.
2. Mirror the actions of that CEO and make sure I remember every single person and what they did; it leaves a lasting impression on people.
A good CEO according to the market, which is what counts in the US, is one that drives up the valuation of a company and increases profit in a mature company. It has nothing to do with personal interaction as most of us would prefer.
Well then you and I differ - I certainly dont turn to the market to define what is good, I guess thats what happens when you worship money, or are operating on the engineers fallacy where you ignore things that are hard to quantify numerically (e.g. empathy)
Genuine question: am I missing some sarcasm here? When you say a CEO has to "talk to people", and then say it can't be done by a chat bot. That's literally the only thing a chat bot can do.
Every person all the way down the chain thinks everyone else can be replaced by AI, but have some good reason that they themselves (or their role) can’t be.
Personally I think that almost any role can be augmented by AI, but very few can be outright replaced that we’re not already very close to an RPA bot replacing them anyway.
Then you consider that they might be able to reduce headcount because of productivity increases. But why wouldn’t they want to use the budget they already have for these newly more productive people and get even more work done overall?
There are some exceptions, like maybe you can be entirely rid of some L1 support.
> But why wouldn’t they want to use the budget they already have for these newly more productive people and get even more work done overall?
For all the same reasons many companies are eager to cut labor costs whenever possible. There can be a variety of bottlenecks to scaling and it's often not as simple as hire more people to make more profit, even if you have extra capital available.
Especially in investor relationships and customer relationships, the talking itself is often not the important part. Yes, you need to align on terms and goals. But the interpersonal component is arguably more important. You need to project that you are a trustworthy partner with compatible goals. And being present as the CEO (or other manager) also shows a resource investment into the future partnership. Your time is valuable, and you took some of that time and dedicated it to this meeting. A chatbot can't do that
Thank you for the backup link! I hadn't known he had passed, and I didn't expect the link to fail. I first read the story back in 2008 or so, and it's been one of those works that lodged itself in my mind so it's sad to hear he died.
I think also it's extremely hard to fairly judge high level decisions whose outcomes may not materialize for a very long time. This makes it genuinely hard for humans to get better (and one gets the impression that most of them don't) with more experience at the helm of a large and complex org. But it also means that it's hard to evaluate or train a machine on the task. You can maybe automate getting something plausible out, but how would you know if it was _good_ in a timely way? So how do you review a configuration or prompt change in this repo?
> After all what's the difference between a bot showing a layoff notice at 5am or an AI generated message/video from a human doing the very same thing.
Of course, any well-trained AI should be aware that the proper way to handle a layoff notice is to send it to your wall-mounted fax machine.
I strongly disagree with this take. AI does not have the theory of mind to manage complex social environments, and AI does not have the context and intuition and experience required to replace a CEO, any more than a random external worker could take over their job for a day.
> I strongly disagree with this take. AI does not have the theory of mind to manage complex social environments
Does Musk have a complex theory of mind to manage his employees, or does he just have a brutally efficient organizational strategy and gets rid of anything that doesn't fit? I think it's more the latter, and an LLM can absolutely do that.
Leadership, Management and being a Boss are not the same things. You need a solid mix of all three to be a successful executive. Cut throat, self aggrandizing, and a bit of a selfish asshole are all traits that lead to positive outcomes in these roles.
The problem I think is that each company usually has one CEO, so automation/higher productivity cannot lead to cutting CEO jobs by 50% or sth. Unless people with AI now can be CEOs in multiple companies at once, as they do not need to spend as much time in a single company now that they have AI? Or maybe with the help of AI they can take also different executive roles at the same time, as in now with AI why should one have a separate CEO and CTO when with AI one can do both jobs at once in the same amount of time one would do only one without AI?
Automating the c-suite doesn’t save you money by cutting headcount, it saves you money by reducing salaries. Having a decently competent AI executive would put a lot of downward pressure on the insane benefits packages these folks get
"I'm not running out of issues. They just keep getting more obscure." Totally agree, For me it's becouse AI takes in places I've never explored before.
In terms of 'automating' Leadership, it's self-evident that Agentic AI is more than capable of inferring context and providing insight based on its significant business administration and management corpus. From this it can suggest actions, follow-up via a plethora of mediums and APIs, and generate up to the minute reporting on a variety of KPIs.
In terms of replacing the function of a Leader, however, the reality is closer to that old joke about the Theoretical Physicist and the farmer - "I have a solution, but it only works for a spherical cow in a vacuum".
The last thing I'd trust to navigate the Realpolitik of modern globalised business is a language model distilled by an Individual Contributor. When Agentic AI navigates moral hazard, it does so as a cosplayer at best - the final actualisation of Dramaturgical sociology.
In fact, Erving Goffman's original Theatrical Metaphors from his 1959 opus 'The Presentation of Self in Everyday Life' map almost directly to the ludicrous symbiotic relationship defined by impression management on both sides of the prompt: AI systems aligning themselves by aping linguistic cues and saccharine affirmations, while users are effectively required to rephrase or reframe prompts to the point of absurdity in order to elicit the desired response, or otherwise navigate guardrails.
Idealized organizations are not perfect. They are perfectly pathological. The top tier are effectively sociopaths operating under conditions of serious moral hazard. If Agentic AI becomes an able replacement, everyone dies.
The only real bottleneck for self-organized discord work groups becoming more mainstream is the need for a central point of contact for legal things, money things, and inter/intra communication things. I see 0 reason why there couldn’t be a system for distributing all of this work at some point.
If you want to add a feature compliant with X Y Z, you need forms A B C and to implement I J K. Put it on the board for whoever’s interested and when it’s finished, add it to the set of things you’re compliant with. If there’s disagreement about how it should be implemented, elect a tiebreaker. Have automatic actions and alerts based on what you’re compliant with and what you intend to do. etc
All this needs to work is the ability to reliably look stuff up by how you want to change the organization, and pattern match on previous approaches and your evolving situations. If you aren’t inventing anything new this is exactly what LLMs are for.
> To be fair I think its far easier to automate leadership jobs despite how they believe in greater self-worth.
I disagree for entirely different reasons than others; I think you can't replace a CEO with AI for the same reason you can't replace a king with AI. It's a title, not a job. You can't even replace middle management with AI - not because it won't (eventually) be able to do the job "just as well," but that the job is largely to take the blame for failure, not to do things.
These people like to take credit for being coordinators, just like property owners like to pretend that that's the same thing as a property manager, but they're different functions. One is a title, and one is a duty. If the person with the title takes up another duty, it's by choice. In the case of both property owners and business owners it's buying themselves a job, often (especially in the case of property owners) because they are in debt up to their eyeballs and can't afford another employee.
CEOs are people who select from and approve choices that are handed to them by experts. You hire and fire them to reassure stockholders. Their only real job is to make complex, convincing excuses for their performance.
Lol!! Don't forget. This is a modern day class war. This current LLM based so called AI, could have easily used for automating simple tasks and employees could have easily provided with 3 day work week. The goal is to monopolize human knowledge with all the stolen data and make profits from it.
The dream of a perpetual labor machine is the end goal of capitalists. It's how you usher in the extermination age. Much easier to do once you declare 20-40% of the population "useless."
> The dream of a perpetual labor machine is the end goal of capitalists.
Frankly, "AGI" just seems like a return to Antebellum slave economics.
Free labor is a C-suite's wet dream, and it has been since the 1520s.
Yeah "AI" doesn't work but that doesn't matter. Your boss believes that there is a coin-operated robot slave that is smarter than you, never gets tired, and never talks back. Good luck getting that raise now, asshole.
The class war is the Marxist equivalent of BS like the "LGBTQ agenda," "white replacement," or George Soros and his plan to destroy Western civilization by liberal policing. Communism and fascism are the two great totalitarian movements from the early 20th century that are still with us, and both run on grievances and conspiracism. That’s another way of saying they run on “I’m mad, so who do we need to punish for it?”
A CEO isn't the boss. A CEO has two collections of bosses, usually. One is investors / the board. The other is the market, which sometimes includes you. The board wants growth, revenue, and profit margin. You want to maximize value for what you pay, so you also want margin, but for different reasons.
So all the CEO's customers are screaming "reduce costs!" including you. If you're shopping and you see two largely equivalent products and one is 30% cheaper, which one will you buy? The cheaper one. So how do you make things more cheaply? You cut costs. For most businesses labor is the #1 cost. Every time you shop on price you're telling the market to cut wages and labor.
That means when you comparison shop, you're telling businesses to cut labor costs. The board and investors are too, of course. A lot of what looks like a conspiracy can be explained by people doing what they have been incentivized to do, and all the incentives say cut labor costs.
Understand human nature, incentives, and complex systems like markets and how they produce incentives. The world will make sense without the conspiracism.
BTW the idea that all the C-suites are all fascists is also false, and is also conspiracism.
There is a loud contingent of far right ideologues in the tech industry C-suite, but this is just the C-suite subset of the loud contingent of far right ideologues that exist throughout the industry at all levels.
The majority of the industry always has and still does skew either left or left-liberal / left-libertarian. The C suites probably skew that way less than the general population, but your average tech C suite person is going to be somewhere to the left of your average oil and gas industry C suite person for example. I've heard right-wing stuff from C suite types, and I've heard conversations at executive/founder/investor functions that sound like Reddit threads about how Trump is going to declare martial law. Opinions run the whole gamut everywhere.
Conspiracism is nonsense. Yes, there are a few people who do plot and scheme and think they run the world. Most conspiracies fail, because the world runs the world. The world is a complex system full of feedback loops, emergent patterns, and incentives.
> Understand human nature, incentives, and complex systems like markets and how they produce incentives.
We do. "Class war" is just the succinct phrasing for "people who earn money through capital returns have vastly different incentives than people who earn money from their labor, and they also have the means to turn those interests into political leverage".
Except that when you comparison shop, you are helping construct those incentives.
When you sell your labor you want it to cost as much as possible. When you buy the results of others labor you want them to be as cheap as possible.
The system is circles, cycles, loops, two sided trades. It’s not one group of people unilaterally plotting things. When someone does try to do that they usually fail unless their plot happens to align with the current in the river.
If you get rid of the investment and capital class you will get a new investment and capital class, probably made up of the leaders of the revolution that overthrew the old one. That’s where our current one came from after all. They are the intellectual and social descendants of the mercantile capitalists who overthrew the feudal order.
It’s pretty common for the new boss to end up becoming the old boss because they are also monkeys with the same limbic system.
Real change would require changing the system not the people in it, but I see no ideas on how to do this from the typical pop Marxist at least. Maybe there are deeper thinkers out there. Usually it’s “proletarian revolution, magic happens here, utopia.” Instead you get a new capital class whose power is rooted in position in the party, like the CCP, and it behaves just like our capital class.
I also suspect really deep change would require genetic modification that alters the structure of the human central nervous system, since a lot of this stuff is just ape behavior writ large.
My deeper point is: there’s a lot of discontent out there for legitimate reasons, but I don’t think LARPing communism will help any more than LARPing fascism has. We need new ideas, not 19th century ones that failed. Right now it just feels like the only ideas we have is to LARP the 1930s.
The people who think everything is a group vs. group war naturally interpret everything as a move within that war, including non-participants' correct observation that it doesn't really exist outside of the minds of the small segment of paranoiacs choosing to engage it.
The concept of "class struggle" epitomizes the "group vs. group war" mentality. It's a mental framework that combines ingroup/outgroup dynamics with a Manichean moral narrative, leading to literally every question being interpreted as a move in relation to an all-encompassing macro-level conflict.
This in turn leads to every problem being abstracted out of its context, and attributed to malice instead of error. Thinking this way completely hampers the ability of human beings to make incremental progress on anything at all via net-positive social engagement with each other.
The "class struggle" idea is of the most debilitating memetic diseases that's ever come about in human history.
Having read history and knowing a bit about labor conditions before workers fought for rights, I find it odd to find someone who rejects the idea that class struggles exist, or that those with riches don't try to protect their exclusive access to those riches more often than not.
Huh. Do you think the same thing about the concept of social/economic class as well? Or is it just rejecting that there is typically a struggle between them?
Basically, yeah. Classification in general is a subjective, value-laden activity. At the most granular resolution, everything is fundamentally sui generis -- putting things into classes requires making subjective judgments about what qualities and characteristics you're going to ignore in order to focus on the ones you care about.
Put simply, when you come up with a class/category taxonomy, you're drawing your own boundaries inside a continuum of variation, and where you draw them is something you're choosing based on the purposes you are pursuing and the assumptions you have made, not mapping out something that's intrinsic to the external world independently of you. "Category X includes object Y" is a property inherent to the definition of category X; it is not a quality inherent to object Y.
And when the things you're classifying are other people, and you're trying to develop a taxonomy that models their behavior, you have a situation where the entities you're analyzing are self-aware beings whose behavior is driven by the analytical models and taxonomies in their minds, not the ones in yours.
People who propose class theories as a usable model for understanding human intentions and behavior lump other people together according to their own criteria, even though those people can -- and often do -- make their own decisions according to completely different criteria. People do not see strangers as having common interests or values simply because a third set of strangers have reduced them to data points that fall into the same bar on a narrowly tailored histogram.
When put to the test, these class theories consistently fail to predict actual human behavior, and people with normative attachements to the theories then to hand-wave away these discrepancies: "false consciousness", "voting against their interests", etc.; these notions are layers of epicycles meant to sustain a speculative theory in the face of falsifying evidence.
Much better to do away with a priori class theories, and just analyze what actual people are actually saying and doing.
So you’re basically against the concept of classes and the act of classification. An interesting position to take in a hacker community but ok :)
However you can’t simply dismiss a whole literature of class struggle by simply rejecting the notion of class fundamentally.
Sure, every human being is unique and individual. But if you take a few million of them, patterns arise. Marketing people know and abuse this.
Also class struggle is not about an a priori theory. If you actually look at what people are doing, you can observe similar patterns too. You can see for instance that billionaires think and act in a specific way, which is widely different than, and against against the interests of, a common working person. There is a fundamental conflict of interest between a boss and their worker. You can observe this and they will actually say this if you ask them.
Yes some groups fail to act for their best interests because they are oppressed. A slave is obeying his master because it’s the best option for him under that circumstances. Maybe he’s not even aware of the possibility of other options. He may even say that his master is a very good individual and he’s proud to be his slave. None of these means that there is no conflict between those two classes, nor there are no classes.
If the categories here are truly a priori, then it seems strange to mistrust them? Maybe you mean a posteriori?
I think I get the angle here, but the whole thing about a single category/classification/abstraction against a continuum is that it doesn't need to speak to or predict the entire continuum, or to every quality and aspiration of one of its members. One set of things, "those items which have X", need not exclude their members from being in another set which might deal with Y and Z! The concept of a single class, whether we are talking Bertrand Russell or Marx, != complete "taxonomic" classification. As such, a social class is, yes, a lossy abstraction taken from the point of view of any given individual, but it's not like it's failing at being a good category because this is so. That's just, like, how concepts in general work?
I wonder what politics looks like following through with your argument? How do we conceive of things like the "public best interest" if any given classification is doomed to not capture people actually? What is the "public" here? Are there children? Seniors? Veterans?
In my experience, these class warfare conspiracists see class warfare in everything. Everything in human society gets reduced down to one side is the oppressor and one side is the victim. And they’re always the victim.
EDIT since I can’t reply:
It’s true that capital and labor often have competing goals, but they also have alignment at times. And labor also has different goals from consumers, from other groups of labor, etc. Capital owners are also in fierce competition with each other! And many individuals in the world are playing multiple roles in the system at the same time! I’m a consumer, a capital owner, a CEO, a wage worker, etc, all at the same time. I also have a bunch of stakeholders that I answer to: customers, investors, employees, the government, etc.
So I object to the simplification of reducing a global system of ten billion humans with their own agency and circumstances, all collapsed down to one side vs another side, with the capital class in conspiratorial lockstep with each other, when the reality is much more complicated and interdependent.
I also object to the constant claims of conspiracy among capital holders. Capital holders are also in extremely fierce competition with each other, and they are in turn beholden to consumers / buyers, etc.
> I also object to the constant claims of conspiracy among capital holders. Capital holders are also in extremely fierce competition with each other, and they are in turn beholden to consumers / buyers, etc.
Cause we’ve never seen capital holders collude to keep wages down. Certainly never in the technology industry. It’s just pure competition as nature intended.[1]
Of course we see all kinds of illegal, anti-competitive, anti-consumer actions and collusions, etc. We also see all kinds of horrible actions from wage earners, consumers, the government, etc. What is your point?
I'm clearly talking about an attitude that basically all capital owners are in secret cahoots with each other to destroy their sworn enemy: workers. Aka, the class war they don't want you to believe is happening.
They were literally in cahoots to keep wages down, you can’t hand wave it away as nonsense when it’s factually occurred.
You do have bad actors amongst labor but everytime labor tries to coordinate together the union crushing apparatus comes out so you don’t tend to see as much coordinated bad activity.
And with AI we’re not seeing companies chomping at the bit to make all their employees more productive left and right, we’re seeing them chomping at the bit to see how many employees they can cut
I've actually had a lot of success with something similar for my startup. I'm a first time founder, and there are so many unknowns when you're getting started. I've been using a Hermes agent to act as my "boss", giving me three tasks to do each day / creating a memory of everything with the business.
It's really, really helpful. So many things I would have completely missed if it weren't for this. I don't think you need this synthetic org setup to get exec help at a small level. It really helped me with fundraising, incorporation, compliance, and just keeping me on track.
Based on what? How do you judge if those are the right things to work on? As a founder, your number 1 priority is essentially to manage that the company works towards the right thing, but it sounds like to me you're outsourcing that part to a LLM? How do you know what the "right track" is when the track is just what the LLM outlined for you?
> So many things I would have completely missed if it weren't for this.
This is the "unknown unknowns" problem. It sounds like the agent is a reasonably good source of obvious-in-hindsight problems that a experienced founder would have experienced, but a new founder wouldn't easily guess at.
> judge if these are the right things
This problem exists no matter where you get the things from.
This is a fantastic idea and I'd like to add my voice to the choir asking you for more information :) Don't let us rebuild everything from scratch if we don't have to
Would love to see how do you setup the agents, I'm not at the level of director but I think it would be good too for staff engineer to manage the office politics
I do this with a Gemini instance that is connected to my email and calendar and task list. Every morning I get an email with an agenda. A lot isn't explicitly taken from my task list, but is instead inferred from the contents of emails.
As a mundane example, I had an email from my wife about a kid's swimming party she'd RSVP'd to.
My agenda that morning reminded me to pack, inter alia, sunblock and towels for my new car's seats. So it inferred sunblock from the party mention, and towels from the party mention alongside emails talking about my new car (and some "knowledge" that people take more care to protect new cars vs old cars).
As an engineer, I love the idea of flipping the problem on its head. Don't get rid of the engineers; get rid of the CEO. That's brilliant.
But if we are seeing problems with agents replacing knowledge workers at the coding layer, why assume the executive layer is uniquely immune?
There was an interesting article in the Harvard Business Review about this exact problem. Researchers found that when LLMs are asked for strategic business advice, they tend to converge toward what is popular and trendy rather than necessarily what is best for the particular situation. They called this "trendslop," and described the response as,
"On strategy, LLMs might be more akin to a freshly minted MBA or junior consultant, parroting what's popular rather than what's right for a particular situation."
Counterpoint: most human CEO's are doing trendslop as well. At least in the worlds I live in (US technology industry, so either megacap public companies or SV-funded startups). It often seems like all of the VCs, CEOs, and large investors in public software companies are all on the same group-chats and sharing the same ideas with each other, driving lots of this trendslop.
From observation of having worked at 7-8 companies as a software dev, it sure seems to me that most CEO decisions could be replaced by AI easier than software dev work can.
Neither CEO nor dev here but must say hn comments on CEOs are always wild. It’s like some caricature - like an attempt to draw an animal that has never been seen from a description
Somehow founders constantly putting out fires, leading and strategizing makes sense but once they acquire the CEO title they apparently do nothing and swan about
The founders who become CEOs are generally the better ones. The ones that draw universal hatred are usually hired by the board of public companies, to enact unpopular measures to ensure shareholder returns, such as firing expensive developers to make room for cheap AI; or unfairly denying health care to policy holders (like Luigi's alleged victim); etc. The relatively rare heartless corporate CEO also receives disproportionate public attention compared to the more common founder->CEO.
But there is literature concensus that at least in the US, non-US born CEOs are the best, and there is little relation between founders -> CEOs and good company performance on multiple metrics.
They can be the best at bringing in shareholder value, which would have a great correlation with enshittification. Having the ability to reduce the quality of something so that customers barely stay (or have no where else to go because they have been trapped in a monopoly or a walled garden) is not a skill. I'd argue it is a trait out of cartoon villains.
> The ones that draw universal hatred are usually hired by the board of public companies
Nah. When I talk about CEOs I usually mean founders-turned-CEO because those are the types I've personally known and shared space with. Remember, this is HN.
Most of those guys are megalomaniacs, clueless, or clueless megalomaniacs.
The U.S. justice system is notorious for pushing people to plead guilty when they really aren't, or for cases involving extenuating circumstances.
The apparent facts bear out that Luigi is guilty of killing that executive, but since it was an executive in a position of making decisions that damaged or ended the lives of hundreds, if not thousands, of people, they would give him a pass (and directions to other executives/billionaires).
I didn't say anything about alleged. I said the "apparent facts". I chose the phrase "apparent facts" because prosecutors and police outright lie or, at best, misstate the facts about what happened, such as when ICE agents claimed that the people they shot were driving their cars at them.
No, he didn’t. Not by any logic that doesn’t also implicate every other executive, employee, vendor, and customer of every health insurance company, plus a huge number of other industries. To the point where the blame is so diffuse that it’s a pointless exercise.
Edit since I can’t reply: if “murderer” means someone who takes actions that have nth-order effects of someone dying, with no reasonable or discernible limit on n, then we’re all murderers and the term means nothing.
So let's hold him accountable for the illegal things he did!
Oh wait, he didn't break any laws? Hmm...well, fuck the idea of having a legal system I guess, let's just cheer every vigilante who decides to murder people in the street based on their own feelings. I'm sure that'll work out well.
Luigi should just done something that caused each of their executives to lose 10 years off their lifespan without actually killing them. Then he would have gotten rid of an entire lifetime but because it’s spread out, you couldn’t call him a murderer.
No, that's stupid too. Insurance companies pay for things that extend life. They don't shorten life. That would be things like tobacco companies. And even then, it's actually the customers of tobacco companies that do it.
This is the same talk that saw people turn a blind eye to all the horror in history. You invent a fake story that casts someone (Jews in Nazi Germany, the middle class in Soviet Russia, whatever it is) as worthy of death, and then you do everything you can to excuse the behaviour of their murderers.
I can tell you that in my case, at least, the real areas of opportunity for something like this come from the masses of people that flood the manager up to senior VP level in a mega-corp. When I'm sitting in a meeting with 8 people and myself and 1 other person are doing most of the work while the rest are vaguely accountable for some metric being reported (we two talk about what's going on, plans, maybe present something completed while the others say "I don't have anything" and the meeting ends, repeat ad nauseum) I start dreaming of what a setup like the one posted here could do. Feels like we could cut at least 50% of our people at those higher levels and stay perfectly efficient and make progress.
Mega corp has too many people who define success by how many reports they have, how many charts they present (prepared by others who are perfectly capable of presenting themselves), how many meetings they sit in and how many reviews they conduct. Automate them away.
I'm not going to vouch for every single HN comment, but running a large established company is fundamentally different from founding a startup, and I think most people recognize that. Getting a big company CEO job has way way more to do with networking and reputation than any sort of technical competence.
When one of those jobs comes open it's almost guaranteed to be filled with someone from a specific .00001% of the population. Who most of the time also happens to be golfing buddies with at least one board member. Are corporate boards that amazing at sifting through millions of potential candidates and finding the meritocratic best fit? I doubt it.
> It’s like some caricature - like an attempt to draw an animal that has never been seen from a description
In my experience this is how HNers see a lot of roles or professions that are not software developer. Finance = Patrick Bateman, lawyer = Saul Goodman, etc.
There are a few different scenarios that people fit into:
* Lack of focus and discipline: Good founder, huckster, good at sales but won't scale with the business when it needs to start focussing.
* Fearful and nervous: Might start a business but doesn't want to make the wrong decision and so stops the business from growing due to decisions not being made when they need making.
* Not a founder but got the role somewhere once, lacks genuine transferable skills. Might work OK in another business but more than likely just riding the existing success/failure of the new business and leaves after 1-2 years to "face a new challenge".
* The rarest of all. Has all the relevant skills, creative, outgoing, mission focused and makes hard decisions efficiently. Will do well in whatever business they are in but might not be very popular!
I dunno most CEOs I have been under of seem to work really hard and try to get stuff done in a sane manner, it is usually the middle management that fs up.
CEO is, of course, partly responsible for having that middle management in the first place. However if I were the CEO I don't even know how I would be able to smell the rats with the top-down view that CEOs have (they can't be in every meeting and decision making of their subordinates).
If you have the right underlings (or inherit them) you might actually not need to do anything at all and succeed. Then again if you have the wrong underlings (or inherit them) then your job is hella-difficult.
Interesting. I have had mostly the opposite experience. CEO(s) that changed strategy + focus every month and the middle management having to pick up the pieces. Keep people motivated and deliver.
I've been in a situation where I had to work directly with a good CEO (plus board, C-Levels, and a lot of middle-management) and problematic middle-management was the actual problem.
It was almost like parody:
- CEO wanting people to leave at 5 sharp, middle-management inventing work and forcing people to do overtime and telling people "he's the CEO, he can leave at 5, he doesn't know we have a lot of work to do".
- Bureaucracy and inefficiency, with everyone complaining about it, but whenever CEO pushed back on that same bureaucracy, the discussion would be hampered by one or two people fillibustering the meeting to force the status quo.
- People going to golf courses or resort trips paid by vendors and then pushing for a vendor that costs 100k/year. This one was so blatant that I was surprised said person wasn't fired.
- CEO and C-Levels taking suggestions from the trenches, and the board loving it, but middle-managers laughing about the suggestions behind the CEOs back, right after the meeting. This one stung because a few of those were suggestions I made and he presented as his own on my request.
This is the kind of stuff that HN people would go "Chesterton Fence" but was just fucking corruption and schoolyard bullying from middle-management.
Obviously the company didn't grow in the period, and the CEO resigned and left for much greener pastures.
In the end the only lesson I learned is that the CEO should have laid-off those people.
And that the CEO should make more money because childcare is hard work.
Yeah this is what I have seen in the orgs I been in too (although not as bad as you pointed out), I have also seen CEOs fail by inaction like you pointed out (not laying off people doing bad things).
Although I am more afraid of the middle manager who hides information than the CEO who doesn't act on it.
I'm no MBA (insert joke here about that being good or bad), but my general opinion is that leaders should delegate and then be held responsible. Details (beyond enough for the higher leadership to be effective) should be relatively minimal. Having some cross-functional accountability (eg security/financial audits, etc) is fine, but in my experience the vast majority of corporate dysfunction comes from micromanaging. This gets compounded when middle management also has to interpret and implement said micromanagement. Obviously people need to learn and make mistakes; mistakes are fine so long as they're not repeated! There is a reason that some of the best, most effective leaders I've ever worked under shield those under them from the politics of the organization.
I appreciate that in the real world a lot of nepotism means that people under higher leadership are the ones to be held accountable/scapegoated, though. Grifting has become a massive problem in the modern corporate world. You see it all the time when failed CEOs get such lucrative golden parachutes.
>qIf you have the right underlings (or inherit them) you might actually not need to do anything at all and succeed. Then again if you have the wrong underlings (or inherit them) then your job is hella-difficult.”
I think this paragraph is right on the money, but the ‘leader’ has more influence on their ‘underlings’ than you give them credit for. In this vein, I highly recommend “Turn the Ship Around!” by Marquet, which is the only good book on leadership I’ve managed to find.
Yeah if new CEO gives the middle management mission in the lines of - we need high productivity, lets be nice and supportive to our teams to get the best out of them since we believe core of our employees are smart hard working folks, it trickles down into thousand little (and not so little things).
Reverse also trickles some brown stuff down. CEO who doesn't care about that leaves stuff to middle management, who are often over-eager to please for the carrot on stick in form of bonuses.
Yeah it is that old saying, the best person to wield power is the one that doesn't want it. High pay and bonuses get rats into the ship, low pay makes no competent sailor want to join your crew.
Not really. I’m not contrasting two peoples opinions on the same thing, but rather aggregate vibe on two different but functionally similar things.
Bit like if hn was convinced white cats catch mice better than black ones. Point out that this makes no sense wouldn’t be a fallacy, even if it involves aggregating opinions from people with their own views on cats
CEOs apparently don’t bother to share their calendars with ordinary workers, and in many of the times when they do interact with them it feels more like lip services, so it’s natural to assume that they don’t do much.
But in reality it must be a hugely difficult job. I can’t imagine dealing with so many humans myself.
Obviously there is simplification going on, but clearly somewhere between founder of a 5 person startup. And CEO of a public multi-national there is a line where they really are not much alike.
I mean, here the submission is about a CEO who fired the company to replace them with LLMs. That would have been a an HBO‘s Silicon Valley joke a few years ago
Both exist. I’d imagine most HN commenters do not work for startup headed up by a founder. The majority work for larger companies where the CEO isn’t as hands on as a startup founder and is never seen in the workplace. I promise you there’s a reason the stereotype exists!
That is true but “is never seen in the workplace” to doesn’t add value is still one hell of a leap.
At scale CEOs don’t add value by clocking in and pumping out LoC themselves and they sure aren’t having all their employees tag along on their day to day.
Reckon there is either an underlying belief that coding LoC is the only value and everyone else is contributing nothing or a if I can’t see it then it must not exist. Either way is not great.
But yeah sure I suppose it is possible that there are CEOs that do nothing yet get paid millions and somehow holds on to a competitive job by some unspecified magic. Doesn’t seem hugely plausible but yeah maybe
In reality the leap for a significant portion of HN goes more like “is not seen in the workplace” -> “is a sociopathic monster who is waging class warfare against me and my friends and all that is good and true in the world, and deserves to be shot in the back on a public street”
People are fucking unhinged. Thank god these people are almost all too lazy and cowardly to act on their little revenge fantasies.
I think a founder and a CEO are two different animals entirely. The two roles couldn’t be any more different, they just happen to share a title.
This caricature that’s occurring could also be thanks to all the horrible CEOs being so visible to everyone.
One of my friend’s medium-size billion dollar-ish company with almost a thousand employees is run by a CEO who is currently in AI psychosis and has started regularly submitting pull requests and demanding they get reviewed as if he’s a member of the development team. He basically comes in with a wrecking ball to carefully planned work with his own daily inspirations that he demands teams far down the totem pole drop everything to work on.
This isn’t an animal that has never been seen, this is a real person that is acting a real fool and negatively impacting my real friend.
We have all been fed this narrative that CEO is the most demanding job in the corporate ladder. How does this CEO have time to be submitting pull requests?
Then you have people like Elon Musk who demonstrate categorically that it’s actually possible to be the CEO of multiple publicly traded companies at once. The fact that this is possible at all smashes the facade of the “impossibly busy CEO.”
Look at how many CEOs have other corporate engagements outside of their primary CEO job. How does Tim Cook have time to be a board member of Nike? How does Satya Nadella have time to be on the board of Starbucks and the University of Chicago?
Logically, if the CEO is such a rigorous job with grueling hours you shouldn’t be physically able to do it at two companies, be a board member for completely unrelated businesses, or run a philanthropic foundation on the side in your spare time.
Then we move on to the "rich guy toys."
How does Mark Zuckerberg have the time to utilize a $300 million yacht? If he's so busy why doesn't he just charter one on the rare occasion he has a week off? Isn't a boat a horrible place to be when you're a busy CEO who needs to constantly travel to company sites, industry and government engagements, and be in close communication with company leaders? Keep in mind, he has owned large scale yachts long before Starlink was a thing.
This is the kind of stuff the CEO of the largest social media company in the world has time for: multiple months of vacationing on his yacht with transponders turned off.
> His $300 million superyacht, Launchpad, has been sailing in stealth mode for the last three months, with its AIS (Automatic Identification System) turned off throughout the European holiday. That’s more than 90 days where the 388 feet long vessel is sailing blind in the busy waters of the Mediterranean.
> Zuckerberg is thoroughly enjoying his time on the newly acquired $300 million luxury vessel. Back in May, the billionaire celebrated his 40th birthday by flying to Panama on his private jet to board the Feadship-built megayacht. A month later, the Meta boss whisked his entire family away on Launchpad on a luxurious cruise to celebrate his father’s 70th birthday, which also coincided with Father’s Day.
Being a non-executive board member often involves very little time or attention, it's mostly lending your name and credibility to the organization. Consider it an American form of royal patronage[0]
10-20 days a year is in the same ballpark and is still a lot. I quoted 6 work weeks (40 hours per day at 5 days per week) which is 30 days a year, not that much higher than your stated 10-20 days.
The article I linked specifically mentioned that private company boards demand less time than public company boards, around 200 hours, roughly 25 work days.
Any of these figures is pretty close to my entire yearly PTO allocation as an individual contributor.
Nike consistently has 5 board meetings per year, so Tim Cook is taking time out of running the second most valuable company in the world to do that, for some reason.
I get it’s a joke, and often a frequent cynical joke made here, but I actually think it’s a good thought experiment. Almost like a new Turing test
The thought experiment, to be clear, is building an artificially intelligent CEO or company founder with the intent of generating profit. If this AI can turn a profit, it has passed this new test
Of course, by the time this new AI starts turning a profit, it will be maximized to the extent that it loses its competitive advantage, and then we’re back to square 1. That is, when everyone uses the same AI, nobody has a competitive advantage over anyone else
My current theory is that because of this, it is not possible to build an AI that generates a profit perpetually. Only temporarily. Then human intervention would be required to adjust things to make the AI competitive again. But at that point, it’s no longer an AI driving the profit, it’s its human tinkerer
This presupposes that a company's competitive advantage comes primarily from their CEO.
It also presupposes that success is zero sum and winner takes all, which isn't always the case (example: when you're building out a new market segment, it can be beneficial to have competitors that are also building it out).
> This presupposes that a company's competitive advantage comes primarily from their CEO.
It’s the premise of the thought experiment: to be able to make a profit, the AI executor must achieve a competitive advantage in the market. If the AI doesn’t generate a profit, then it doesn’t fit the premise and that’s a different experiment
I suppose you can claim that a profit can still be generated by offering a holistically worse service/product with no single advantage than a competitor because markets are not efficient, though this may be short lived
You’re presuming here that CEOs cause companies to make profit instead of serving as a scapegoat for their often unavoidable market dynamics (at best).
I may be able to do all the labor myself, but need investors because I can’t afford the capital.
I may be able to do all the labor myself, but want to ensure continuity beyond my ability or interest to do the labor myself. (Sometimes this is by my wish; sometimes by the wish of consumers of my service.)
I may be able to do all the labor myself, but not want to expose my entire net worth to the financial obligations of the company.
There are plenty of cases where a single person company makes good sense.
At what point does a company become a company if one person doesn’t make a company? What’s the threshold for non-labor where a company becomes a company?
Yep. But, in most senses, trading algorithms don't need to interact with humans, or human society. Is the world around the "thought experiment" AI CEO dehumanized enough that that's also the case for it?
If not, I'd expect major problems there.
A related point - trading algo's can completely pivot in seconds, and live & die by extremely simple and short-term metrics. Even in a human-free world, that's not at all how the vast majority of industries work. What is "profit", when your AI CEO is organizing the construction of a new steel mill? Judging its performance then requires deep domain expertise.
Doesn't the incentive here, lead to privatizing the AI and possibly releasing a different product to the public - or at least with a delay? (not too unlike how the frontier labs work today).
Yes I would think that would be the endgame goal. But in practice, I’m very skeptical a purely AI-driven executor can stay competitive when the market and government supporting it and surrounding it is driven by humans
A serious point about this is that it's an AI that's an organization. This matters.
People laughed at Gas Town, and at that "Fences, not Sandboxes" article yesterday.[1] They're both AIs set up as an organization, not an emulated human. They're both expensive to operate, because the AI members of the organization spend a lot of time talking with each other. They're both applied to frivolous problems.
Now we have the same concept, applied to a more useful problem. It's time to pay attention. Something important is happening here.
Everybody has been talking about "human-level AI", or "superhuman-level AI". This is a different kind of AI - corporate-level AI. It works and thinks like a company's management. It watches a wide range of data sources and responds to some of them with action, and some with internal discussion within the AIs. It can potentially pay attention to far more inputs than any human. That overcomes a basic problem of corporate organization.
A classic management failure is not paying enough attention to something.
This is usually not a problem for computer systems.
Humans are bandwidth-limited. Teams of AIs are far less bandwidth-limited.
They may soon be outperforming humans for that reason alone.
Think about this for a while. These team AI systems may be very useful.
You're probably totally innocent, but this comes across as being AI-written, especially in the context of an AI problem and when your response is pro-AI. For the next 3-12 months (maybe more, we'll see), I suspect you'll have a lot more success if you shy away from those speech patterns.
Because it's only half true. Yeah we've used phrases like those before, but if you ignore certain very specific phrases or characters for a moment, a lot of the more ambiguous things people are identifying as AI-speak comes from mixing jargon from different domains in a way humans normally don't do. Like the short punchy sentences are from marketing, not comments or technical writing like they've started showing up in.
Or at least he's being unintentionally influenced by it. Just out of curiosity I jumped a few years back in his comments looking for that pattern, and I don't really see it.
By pure luck I even landed on a comment that uses that same word with a similar sentiment, except back then he wrote it as a full non-punchy sentence: https://news.ycombinator.com/item?id=31414288
> Heh. We are now at the point where people might end up deliberately inserting typos and bad grandma into their writing, just to signal they aren’t AI.
Except what humans write will continue to be fodder for the LLM's so they'll "learn" that at some point.
The Anti-Turing Test - Prove you aren't AI becoming a thing is very... well Black Mirror.
It's been weird to see the face of a bunch of youtubers I've followed for years without any idea what they look like (mostly history/military history/engineering type stuff), it used to be video/photo's/diagrams with them providing a voice over but because of the sheer quantity of absolute shite/slop mirroring that style they've started appearing on camera as quick "proof of not AI" check.
For that type of content YT is essentially completely ruined if you didn't already follow the good ones because finding them in the absolute sea of dross has become way more difficult.
A friend showed me a video, put out by a local news organization, that started with an attractive woman, standing in the place being discussed, speaking into a microphone.
Cool. Standard local interest story.
Then, it cut to a photo montage, with “her” voice, continuing the story.
Except it was AI. It wasn’t especially successful. You could hear the “tics” from the AI.
It's long READMEs with 3 punchlines per paragraph for a project best described has "I found these 4 thingies, all of them had compromises, so I told Claude: merge the 4, no compromises!"
And every issue, every PR you open, about every small improvement or new feature, is the same.
Dumb people want to impress by using words others might not understand.
Smart people want others to understand them.
In early days of Internet, it was a domain of smart people.
And native English speakers have no problem communicating with people from Ex-Comecon countries with vey little English knowledge.
Hmm it's being done in a less cheesy way than AI does it.
AI text feels more like "this matters, not less." There's always an "it's x not y" pattern somewhere.
It could be written AI assisted but then look at his account. IMO every HN account that was created before ChatGPT came out has incentive to be written by a human since they have a paper trail before the whole LLM thing happened.
This is at least written by a human, not AI, totally human, (digital) pinky promise ;-)
Like hell we can't. Unsophisticated plebs (not you) viewing sophistication with suspicion is nothing new. And certainly no reason to demote one's standards.
Long live the em-dash. Or en-dash surrounded by spaces if you prefer UK style (which I do).
Sigh. I did write that personally, just as I have all my other HN posts. I'm moderately well known, and not anonymous. See www.animats.com.
I wrote that because the last two times AIs built like organizations came up on HN, they were dismissed. Hence the somewhat out of character rhetoric.
We have organizations of people for the usual reasons. Scale. Specialization. Robustness against failure and errors. All those wins apply to structured groups of LLMs. So this has a reasonable chance of being useful. It's expensive at the moment, but probably cheaper than the management team it may replace.
For some reason, this idea seems to upset people. Unclear why. Comments?
FWIW, I thought it was human written. Pangram also said that it's 100% human written.
More likely, they're all wrong on this accusation. A bit concerning to see how many of them are throwing out a false positive like this. Makes me wonder how many of them are falling for actual AI traps in the real world. :)
I am actually with you here. Such a system probably needs to be battle tested in a specific domain of business to see where the deficiencies at this stage are.
Problem is, no matter how many sub-agents you break the LLM into it's still stochastic parrots all the way down. I've seen no evidence that compartmentalised harnesses can reach solutions that a monolithic harnesses cannot, so I'm thinking the compartmentalisation is just an excuse to throw more tokens into the token furnace - it doesn't unlock a step change in capability. If there is evidence to the contrary, I'd be interested in seeing it.
There have been some successful attempts to do science like this (https://github.com/AstroPilot-AI/Denario). Denario generated 1k+ "solutions" on its way to winning the FAIR Universe challenge, and they all had to be fed through a fitness function of some kind to assess them. It also got stuck on a local maximum and had to be kicked in the ass by a human to get it off that.
When writing software, that fitness function is relatively closed (when I run it, does it do what I want?). With something as open-ended as managing a business... might as well roll dice, IMO.
My belief is that 20's LLMs have a lot in common with 90's GAs: the more clearly we can define "success" for a given task (the fitness function) the more successful they can be. Open-ended problems are somewhat beyond them right now, and, I suspect, forever.
(The comment that's currently at the top of the discussion, about making an "AI Boss" that remembered everything about the company, and fed the user three tasks a day? That's not an AI Boss, that's an AI Assistant with different framing.)
I read the line and thought it was AI, and then saw the username and figured it wouldn't be. I find my writing being influenced by LLM patterns as well, because that's basically all the writing that exists these days. I don't like how it erodes my own voice because I keep reading the writings of this one single author all day, every day.
Well, you mentioned Gas Town — that deserves an immediate reflexive downvote :P
I actually upvoted your post. I'm not sure I entirely agree with it, but I was interested to see more discussion about it, and that felt worthy of an upvote.
As for the reaction against the idea, I see it from two angles. Is the standard corporate hierarchy / architecture really the optimal organization principle for AI? Or is that just a skeuomorphism to make this project seem more serious than it really is? Maybe Yegge did us a favor by mapping to Mayors & Polecats, forcing people to question if it really is the correct organizational hierarchy?
The other angle is heresy here - HN is full of employees now. There is little incentive for employees to be interested in what a C-suite does, and we see that with the comments about CEOs needing to be six foot and walk impressively. Employees have no interest in AI organization principles that optimize them away and highlights their obsolescence. I am surprised there hasn't been more discussion about how this specific project optimizes away the C-suite humans instead - I would have thought that could have had some appeal to employees, at least until those people realized it makes employees a swarm that works for AI leadership.
As for AI organization, the organizational mapping needs to be productive, not affectation or cargo-culting. That's not peripheral — that's load-bearing. Especially in a corporation model, you can find the organization layers only create busywork with lots of unnecessary middle managers. You don't want your AI C-suite swarm burning tokens on endless meetings about what should be in the upcoming meeting... especially not at Fable API pricing.
It saddens me that we're banning speech patterns on HN, not discussing arguments. I have seen this in another thread yesterday(?), where somebody was accused of being an LLM based on speech patterns and downvoted, no further evidence needed. This site used to be above that level.
It’s not about the LLM usage. It was such a dumb construction:
> <totally vague statement>. This matters.
Like, just because you say something matters does not make it true. Explain your case and then make predictions about the implications of it.
The entire thing was a vague post. Just “this matters” and “pay attention”. It’s trying to create a sense of FOMO but not spending any effort to explain why we should care.
The site used to not be under constant assault of low effort LLM submissions and low effort LLM comments. That people start getting cranky about that is just normal.
I mean on the one hand it is annoying because everybody has speech patterns, I like to say "I mean" and "on the one hand" a lot. On the other hand it seems a sort of self-protective recoil, nobody wants to spend a lot of time talking to an AI. Our jobs and society already make us spend way too much time and effort doing that!
On the other hand: so, people throw the baby out with the bathwater, which is a problem, but not sure if it's as big a problem in the old days when you threw out the baby with the bathwater and it was only one tub of bathwater, because in the present time it's 2000 tubs of bathwater, and the baby is drowning anyway.
I say this as a guy whose editor is complaining an article I'm collaborating on, about how the continual whining about AI tends to have a lot of false positives and is anyway annoying, is taking too long.
Your comment has three hands, that's the kind of collaborative communication style we should all aspire to.
Personally, I suspect good writers will adapt in due time, not because they don't want to sound like AI, but because they don't want to sound cliche. Even the ones stubbornly holding on to their em-dashes.
The ones hit hardest will be the mediocre writers, who have always relied on cliches. Their writing will become indistinguishable from reading chatbot output.
> Your comment has three hands, that's the kind of collaborative communication style we should all aspire to.
The Zaphod Beeblebrox style of communication.
So far I haven't been accused of using AI for my comments, even though I use (and always used) a lot of the tropes. It's probably because my rhetorical skills overall suck.
It’s the whole internet, it’s not an HN problem. The AI industry is little by little invading and destroying every space where humans gather, removing everything that made genuine human interactions interesting and worthwhile.
(Also, don’t take HN votes too seriously, being downvoted doesn’t say much)
Not in the slightest. I didn't touch the voting functionality and tried to be as nice as possible. I saw an opportunity to make somebody's life better for reasons it didn't seem they were aware of, since their speech seemed similar enough to certain AI idioms that I thought it would be demonized, and the fact that they wrote it anyway suggests they didn't know.
As I'm looking at it right now, there might be some nuance I'm missing in how HN works, but the post doesn't seem to be greyed out or anything, suggesting that if it is being downvoted then the upvotes are counterbalancing sufficiently.
> banning speech patterns
AI is prevalent and often bad. Society is reacting quickly but maybe not quickly enough. Blanket speech "bans" are part of that. Is that bad? Eh. It's not ideal. It's to be expected though. I work with the society I'm given, not the one I should have.
A related point that I've been making privately for at least ten years to anyone that would listen, is that we've had Superhuman Intelligence for at least hundreds, if not thousands, of years in the form of corporations and institutions.
No single human could design a rocket to go to the moon or keep all Wallmart stores supplied on a daily basis. Moreover when something like the CEO or US president changes, the policies and actions change a bit but the majority actually stays the same. This is collective intelligence embodied in the organization and its institutional memory. This is also what i believe is wrong with Searle's Chinese Room argument - the algorithm is the intelligence.
Our cells and neurons act based on local chemical gradients and electrical signals and we wouldn't call any single one of those particularly intelligent. Yet collectively they produce human intelligence and even what we call consciousness!
I've made that point before. From an "alignment" perspective, AIs and corporations have many of the same problems. Corporations, according to Milton Friedman, have the sole duty of increasing their profits.[1] This is a key document in the history of modern conservatism.[2]
So what happens when AIs are "aligned" purely to corporate goals? This is a nearer term worry than machines taking over for the benefit of machines. AIs working hard for the benefit of corporations are already here. Try any customer service chatbot.
There's reluctance to think about this too hard. It raises embarrassing questions about the role of the corporation in society. It can become a challenge to corporate personhood.
That's threatening to many people in power. Dealing with it may be unavoidable as AIs take a stronger role in managing companies.
Also the accountability problem! Sometimes we try to hold the corporation accountable and sometimes the individuals. It never seems adequate though.
I found the documentary "The Corporation" (2003) very instructive in that regard, particularly the segment with the BP CEO where the point is made that a company made up ethical individuals can still act unethically, or unaligned to human values, just by virtue of the incentive and legal structure governing the corporation.
“ This is also what i believe is wrong with Searle's Chinese Room argument - the algorithm is the intelligence.”
“ and even what we call consciousness!”
This misunderstands the point of the so-called Chinese room thought experiment. You want to conclude that consciousness is functional, but you cite a thought experiment which argues it is not, and then assert the thought experiment is wrong because it disagrees with your preconceived notion of consciousness.
I mostly agree with you, but referencing Michael Levin, I'll contend that even individual cells have quite some intelligence. Granted, different from ours, but really good at solving the kind of problems cells have to solve.
Agreed. Michael Levin's work and discoveries are mind blowing! I was actually referencing that with the "electrical signals" nod (as well as neuronal activity) and Alan Turing's reaction-diffusion theory of morphogenesis with the chemical gradients. I'm not a molecular biologist so i don't know anything about the details, but the emergence of global phenomena from these local interactions fascinates me.
There's no intelligence without teleology. Intelligence and information are only meaningful concepts of you assert (necessarily without any evidence except the circular kind) that some differences matter.
Do the cells "have" to do anything? Do we? Sure, they might die if they do the "wrong" thing, but 1. Who's to say the inert state of "dead" matter is any worse than the state of a living cell, and 2. Cells clearly aren't always optimizing for their own survival anyway
Intelligence is only meaningful with respect to a goal or set of goals.
I think organizations' intelligence can only be called superhuman if you ascribe them their own inscrutable goals, different from what people want their goals to be.
But why would you do that?
It's true that a lot of the debate about AI alignment - the good parts, anyway - also apply to human institutions.
Because that is what actually gets pursued. The set of effective goals an organization is distinctly different from those of any of the individuals in that organization.
You're missing what I'm trying to say. There are no goals in nature. Agency and goal-directedness is something we ascribe to things.
The set of things an organization ends up doing may be distinctly different from those of the individuals, but that does in no way oblige us to ascribe agency to the organization on its own.
You're still not getting it. "Entity", "survival", these are still arbitrary or circularly defined terms. You decided what an entity is (i.e where one thing ends and a different thing begins) and what the essence which "survives" is (i.e. why it's a big deal whether there's a different chunk of nucleic acids similar but not identical to my own floating around 200 years from now).
These ideas are not wrong, but they are teleological. They are about the purpose of things. And nature doesn't come with purposes, it's all something we impose on it.
collective intelligence is not a superhuman intelligence.
Markets solves the resource allocation problem, but it doesn't and realistically cannot be called intelligent (even tho it's been called wisdom of the crowds), in the same way that darwinian natural selection cannot be called intelligent design.
Yeah, wasn't claiming that it was particularly novel or divine insight, otherwise I might have tried to publish a paper about it. I was just recounting my experience on this matter.
> we've had Superhuman Intelligence for at least hundreds, if not thousands, of years in the form of corporations and institutions.
It’s true that collective groups of humans can accomplish tremendous things, but I hesitate to use “superhuman intelligence” for corporations, governments, and institutions.
So much of what they do ends up being very focused on process for the sake of process. And is often so very dumb.
They're de facto superhuman. Even with the friction and stupidity, they still have more agency than an individual human and process a much bigger entropic footprint for both information and physical resources.
You occasionally get Great Leaders who build these systems as personal affordances. But even then these system are far beyond any one person's ability to manage the details. And they often take on a life of their own.
Our individuality is actually a myth. We're more like a colony organism that self-organises into different sub-organisms. The sub-organisms compete with each other, which is why we're not a smart colony organism.
If individual intelligence was aligned along colony goals at shared planetary scale we'd be something else entirely.
The biological systems that support human intelligence are notoriously messy and work together only tenuously with a great many failure modes, including "process for the sake of process"
> A classic management failure is not paying enough attention to something. This is usually not a problem for computer systems. Humans are bandwidth-limited. Teams of AIs are far less bandwidth-limited.
That's a feature, not a bug. CEOs should be focusing on the key goals of the business at a high level.
Having an inexhaustible AI CEO that can pay attention to every little thing in the company could easily turn into a micromanagement nightmare.
"I noticed your key pressed were down 27% last week and your agents sat idle 3 nights in a row. And honestly? The thing to hold on to is that it's not a small thing, it's a load-bearing reduction in your output. Let's try to minimize the blast radius with some belt-and-suspenders productivity tips from the bullet point list below:"
Some of the non native English speakers on my team have started to use “blast radius” and “shape of” in the course of casual conversation. It’s unnerving.
To some extent, the AI may have adopted some of the expressions because they were in its training corpus enough (ie. People used these expressions before). Might be attributable to Baader-Meinhoff phenomenon instead of people copying AI.
I have use both for years. Elixir and functional languages talk about the shape of data. and blast radius has been used at the last 7 companies I have worked at when talking about impact of a negative thing, usually an incident/outage.
The expression makes sense (a not dissimilar "trickle down" got accepted...). But used like in the text above, "the blast radius of your reduced productivity", it better be satire.
You again anthropologizing AI. AI does not have to make everyone busy to look useful like for some human managers, when right priorities are set.
Imagine a manager who knows that company needs in a long term support person even if it's not busy at the moment. Because AI has data of historical precedents and actual calculations. And AI is not acting like a manager who wants to fire you because of gut feeling or possible stock options. Imagine that AI CEO is owned by employees, actual humans who perform the actual work and who benefit from good CEO decisions and most importantly the AI CEO which does not gave to be paid gazilion dolars just to burn the company to the ground and leave with golden parachute. I think there is a good insight that AI has very high bandwidth, that human management does not have, no more 2 week meeting marathons just to get everyone on same page and to get all the inputs.
> And AI is not acting like a manager who wants to fire you because of gut feeling or possible stock options. Imagine that AI CEO is owned by employees,
You’re just describing a company owned by employees who manage themselves. The AI isn’r relevant to your story other than being a proxy and note keeping apparatus for them.
We have some interesting evidence from AI agents. People used to make the plan with a stronger and more expensive model, and then use a cheaper model to do the "grunt work" of implementing it.
But it turns out it's better to do it the other way around. A good plan is easy, implementing it is not.
Really? I'd love to read some write ups on this because I'm still doing it the old way. I was going to say that corporate experience runs counter to this but maybe not; you think the architect is more important than the developer, but actually there have been shifts over time to recognise ICs more and keep them in their roles while managers are more generic. Interesting food for thought!
On a serious note I think that this fact is not understood by many.
There is no such thing as common sense without shared assumptions. So if those shared assumptions are incorrect or not given at all the output of the models will be nonsensical. In order to truly act human there is a need to experience humanity from the first person.
This is why I think technology will get more and more invasive. Meta glasses now, personal agent devices later and neural implants after that with data sharing enabled.
I hope legislation is never manipulated enough to allow an AI to form a corporation or become it's legal director. With whom would accountability rest? Who would you throw in prison if it acts egregiously enough to warrant such a measure? If it's bots all the way down.
this is already possible using common corporate tax & trust strategies. see Wyoming Asset Protection Trust (WAPT), purpose trusts, donor advised funds.
>With whom would accountability rest? Who would you throw in prison if it acts egregiously enough to warrant such a measure? If it becomes bots all the way down.
Corporations already exist to limit liability and accountability. This would be an extension of their current purpose not some weird deviation from.
The funny thing is Decentralised Autonomous organisations (DAO) under Ethereum and other cryptocurrencies had similar goals originally. I do not know where they stand and if ever a useful DAO was created.
If nothing this is a useful demo in real world if used that will show that there are good reasons why hierarchy forms anyways and how power law is at the center of it. Too much democracy may not work in some instances but there may be many cases where the values can hopefully be aligned as well as be lean enough to adapt
Your point is well taken, but one thing stood out to me.
> A classic management failure is not paying enough attention to something. This is usually not a problem for computer systems.
This is actually my #1 problem with frontier LLMs. They are plenty smart, but they have no true long term memory, and their context window is not nearly enough for a complex system like a corporation. When my project gets big enough and I'm using a lot of /compact, the LLMs always have the same failure point: they fail to pay attention to something.
When they figure this out, I think it's going to be a very, very big moment.
Meanwhile, I have the opposite problem. An errant word or phrase and they hang onto it forever, completely corrupting all future work because they can’t ignore irrelevant (or wrong) details/errors of their own design.
It’s called using systems of record. I use redmine and discourse. Jointly with my stack of agents (COO) (managed by four very high trust humans via OWUI interacting with scoped Hermes profiles ).
I’m adopting this stack under me (CTO) and integrating it into agent zero
Eh, it’s nothing new in general. My software development (both at work and private) is being done by teams of semi-antagonistic agents. I only get to see code after the grumpy reviewer bot ran out of reasons to be grumpy and the pedantic formalist ensured all specs are still adhered to. Saves on the most expensive resource.
Watching the demo video, they haven't created open executives. They've created open managers.
It looks like most of the work of the package is reactive -- monitoring external things and responding. It's not proactive just because a person didn't dial it up.
But, here's the thing -- this can likely replace many executives right now, at least those who are reactive managers. I'm in full support -- automate those things if you can.
AI is coming for judgment jobs. It seems it's turtles all the way down.
Sometimes the claim is "set vision" - which AI can extract from the collective.
Someimes the claim is to prioritize the work (with bias), AI can extract that from the collective too, right?
Making resources available to the team? AI.
Coordinating between teams? AI.
Can the AI do this in a more un-biased way? Can it refine collective intelligence and desire into objective goals?
More than a software factory id love to see a manager-factory.
One thing, AI cannot do (yet) is bring the network. Like when you hire folk from Booth or Sloan so you can sell to places that hired Booth and Sloan grads.
Be friends with the board. Negotiate high pay and benefits (for themselves), and a golden parachute in case things go south (which they do most often). Take credit for things that were going to happen anyways. Blame xyz for things that didn't work out.
These are all impossible tasks for AI. AI can't play golf.
It’s going to be funny when the Tesla robots start showing up at tee times because the CEO couldn’t make it. Like a meeting agent taking notes for people who didn’t attend
See, the issue is CEOs capture a fair amount of value but ultimately it is the owners and the main shareholders who hire them for that value to insulate themselves from day to day running of the company. The maximum value of the company is accruing to the shareholders in a public company anyways. While the morality of the right nos compared to median 'worker' can always be debated, I still find CEO earns 500x that of median worker type comparisons a little bit disingenuous and it totally insulates the owner class.
> The job of an executive is: to define and enforce culture and values for their whole organization, and to ratify good decisions.
(And that's the whole list! Nowhere do you find set tasks or make decisions)
To me, that means that they're still pretty essential - on both fronts, AI needs values and AI needs a system that can tell whether decisions are good or not.
If the operation is to distill options into a few choices, and then pick the "best" one then, I think hivemind can win there too. Real-time democracy running a business.
For example an open-source project could/would still have a BDFL but, the AI-CEO could do the operational busywork while the BDFL is out talking to humans and presenting at PGCon or some other productive thing.
Nope! that's the subtlety. "distill options" is not the CEO's (or any executive's) job.
The way he phrases it, if I remember correctly (and this is by no means the only way to executive, I just really like his framing), is that the way you make decisions is by getting your immediate subordinates to agree, and they iterate. Decisions should be made at the lowest level that they can be made at, and if that's the CEO level, you've made mistakes.
So you make the values list, you make sure that people live by it (up to and including firing people who don't and hiring people who do, which he basically thinks is the whole day to day), and then you bring your subordinates together to agree on decisions. If they don't agree, you send them off to work through it together and come back with consensus (or, less great, compromise). If you have to put your thumb on the scale to get a good decision, again, you're not CEOing well.
Basically if you have to be a dictator, there's no way you can actually effectively do that. Proper leadership is getting people to act the way you would, if you had all the knowledge and perspective they do, but without having to make them do it. Lift them up, empower them, give them the responsibility and tools and values, and then endorse the result.
Business actually treats a pseudo-dictatorship (with the requisite yes-men) as "good leadership".
Most times it's pretty simple, in my experience. A CEO either runs a business as a (1) Sales focus [best deals and vendors] or (2) Marketing focus [best reputation and trends] or (3) Investor focus [best boardroom friends].
Lots of other opinions are usually thrown out until there's really only one facet used for running the company.
Enforcing culture is not something I believe is possible. Nobody is going to listen to the C suite suit over their own experience with their colleagues. People just end up saying the right buzzwords to appear that the culture has been adopted, but it will be mostly met with malicious compliance. If the culture is driven by incentives and rewards for behaviour,then that's a whole different ball game.
At my current job, a 10 person startup, the CEOs job is to generate leads, talk to potential customers, and close deals. He, along with the other founders, shape what we work on based on what he thinks it’s possible for us to sell. He’s essential to the business.
At my previous 50k person company I never talked to anyone within 4 levels of the CEO and thought he was useless. But realistically he probably had a somewhat similar job - talking with the board, talking with politicians, and talking with the CEOs of other companies in the effort to lower costs (tax breaks, partnerships, etc) and generate sales (contracts, partnerships, etc).
Similar, yes, but instead of finding out who else can give you money at 10 people, they have to find out who else they have to give money at 10k people. Leads and lobbies.
A sales person is usually trained on an existing product or portfolio of products (including services). A good sales person might use their exposure to customers for gathering valuable information to feed back into their upstream org. But this is completely different from someone who has to actively shape a both sides of the company, value delivery and value creation, because both need to mature in a choreographed manner.
So you are right in one direction: a good CEO often has to be a good sales person. The reverse does not apply, though.
There’s a reason that many CEOs have a stint in sales.
In a stable, profitable company, you need sales to keep the business solvent. And understanding the sales pipeline is pretty essential to effectively running a business.
Where this breaks down is when the C-suite forgets that you also need a product, and you need to keep innovating on that product in ways that bring value to current and future customers. This seems to happen pretty frequently.
In companies large enough to have a board, the CEO's job is to manage relationships with customers, vendors, and wall street. If his company is big enough, he'll need to manage relationships with the government too. In fact, among founders whose business became big, the number one complaint you hear is how little of their busy day is spent on their actual product anymore.
AI is not very good at forming and building human relationships.
> Sometimes the claim is "set vision" - which AI can extract from the collective.
To be an effective CEO, you have to make judgement calls that the collective can’t always have wisdom and context on, is confidential, or requires fast decision making during which there are interactive discussions or negotiations!
This is highly subjective and a gives an impression that every one is doing a very effective job. The reality is much more grey when you account for so many companies that get shut down.
My personal take, somewhere down the line people feel scared of taking responsibility for more than themselves and their family. This limits the ability for most people to take on such ventures. Not everyone falls in the same bucket but a lot of leadership in todays date is about taking the role without responsibility. They happen to be in the role with the help of network affect and somehow that just amplifies.
> This is highly subjective and a gives an impression that every one is doing a very effective job. The reality is much more grey when you account for so many companies that get shut down.
I believe that the average are average.
I don’t believe what I wrote suggests that. I was attempting to be as neutral as possible.
Good ones actually are quite aware of the industry, connect the dots and setup the company to be sustainable, growing and ready when its time to be ready.
Like investing in a prototype which solves industry issues the CEO understood.
Same as the CTO just that the CTOs focus is doing the same thing with technology. Like in an org with a lot of 'AI? who cares" saying and seeing the future of it and intorducing it in time and not until everyone else already did it.
Defining core principles like the tech stack, making the decision to keep the current setup or migrating to something 'new' but better in longterm (like migrating from or to kubernetes)
Marketing and they hire and fire important people. That's about it though, where we should take the product, what features to do next, partnerships, these are pretty obvious to anyone paying attention.
I don't think these things are obvious. I've worked at many places that have had big, company-changing forks in the road available to them and navigating those forks wasn't easy (and in my opinion several times we chose wrong). The crowd wasn't always right or even well aligned and sometimes the only agreement was that a middle of the road approach would be worse. My intuition says getting that right takes a lot of luck, good momentum, and a lot of departments working well together (bringing in and communicating market info, broadcasting marketing that works, making good partnerships, etc). The cynic in me thinks one big impact of a ceo is the various cheating and corruption tools that can make some of those things easier or at least less critical to get right (you don't need to make good cybertrucks or convince people to buy them if the government agrees to buy 450k of them as "armored vehicles", for example).
If that kind of thing is easy for you (the honest ones or the clandestine ones) then you might make a really strong entrepreneur/business leader.
Yes, that is a good manager, but in my experience, it is better if the culture of the team is about ensuring everyone has opportunities to grow, build on their strengths, and connect with opportunities. When that is considered the manager’s job, and not the job of every member of the team, people get neglected, and many strengths and opportunities go to waste.
They siphon away as much money as they can get away with , screw over their underlings and hop onto their golden parachutes whenever shit hits the fan, which with their short-sighted coke-fuelled decisions always happens.
Not yet. Maybe soon. I'm not placing bets on timelines.
> What do CEOs do
No clue. I'm not there yet. Maybe it's critical. Maybe nothing.
> and other managers
There are a lot of bad managers. I'm halfway not terrible. A significant part of my job is setting technical direction in the same way I did as a staff+ engineer but without the overhead of asking somebody else's permission. Another large part is the "shit shield" function -- somebody thinks they want something, and they either find themselves blocked by me or immediately satisfied with whatever code I'm able to cook up between meetings, not breaking the team's concentration. A huge part is basically just being a teacher, and I love teaching (if it weren't so pessimized by society that might actually be my career). A large part of it is navigating organizational nonsense (for "reasons," I'm rambling, ask if you're curious). And... whatever. My job didn't get easier going into management. Not all of it should have to exist in the abstract, but in any specific company it's still an important function.
The answer is to be responsible to the business. The buck stops at the CEO. Therefore, their responsibilities are technically everything in the entire company. And I think many people who have only worked for other people underestimate how valuable that responsibility is.
Many people observe a limited subset of reality. It's highly visible when a company hits a crisis and people lose their jobs; it's much more common, but much less visible, for a CEO to get an after-hours call about a brewing problem which they must immediately work to smooth over.
I'm not quite sure why you ask that? I'm not arguing that these two very different examples are the same, I'm explaining more concretely what it means to say that a CEO's job is to "to be responsible to the business".
It’s amazing how many people on this website harp on about nobody noticing their hard work, nobody caring unless the server is down, etc, and not being able to conceive of the remote chance of this same phenomenon applying to other roles in an organisation.
CEO's exist as a sink for liability. When something goes wrong, they are the fall guys, that's ultimately their job. That's why CEO compensation is generally correlated strongly to earnings and they usually have golden parachutes.
Without a CEO to fire when the business goes south, how is the board supposed to respond? Taking accountability? lmao
If you were legitimately suddenly saddled with the responsibilities of a typical ‘CEO’, chances are you’d shit your britches within a month.
Part of a CEO’s job is to be the target of all warranted and unwarranted blame for every single bad thing that anyone in the organisation experiences, even the inherent/inevitable ones, even the ones that others in the organisation see as a good thing.
Me. Not a CEO, and no desire to be. I’m also about as lefty as they come, definitely not a bootlicker or ladder climber. I’m just not a delusional, naive, unjustifiably self-righteous grunt, seemingly unlike many people sounding off on this thread.
I deliberately stopped short of invoking golden parachutes and the numerous other escape hatches available at that altitude, as I am perfectly aware they exist. I am equally aware of chief executives who have departed with remarkable agility from organisations they themselves had first steered off the cliff.
My point, however, remains the same: as the letter of the law presently stands, AI can't be made accountable – at least not today.
LLCs and other corporate structures only (mostly) protect employees and shareholders from civil judgments. CEOs and other executives are also at risk of criminal penalties for Sarbanes-Oxley (SOX) violations and similar. Actual criminal prosecutions of CEOs have been few, but not zero.
The point of a CEO is to be able to deploy a fully authoritative business delegation to any golf course on earth within 48 hours.
This isn't really a joke. At some point, you need to meet with your customers if you intend to close a deal. B2C is also an option but good luck actually making realistic income in this market without an army at your back.
I have yet to meet a developer (myself included) that I would trust to be alone in the board room with a very large client. Total mastery of the technology is great, but you also need to convince people who aren't compelled by rational technological arguments. Sometimes the only way to convince them is to go on a very expensive fishing trip.
CEOs get a lot of hate for the pay and power, but they endure a lot of shit that most developers couldn't handle for more than five minutes. Sitting at fancy restaurants with these people is torture. I can respect anyone who refuses to believe this, but it's true. There's no cuisine or vintage of wine that can compensate for awful company.
Ah yes, the unbearable burden of sipping fine wine at a Michelin star restaurant with a less than stellar company.
That's surely way worse than the fate Bezos inflicts on his workers, such as forcing them to work in extremely hazardous warehouses and to pee in bottles.
The crux of this entire issue surrounding replacing people with agents, is that agents should pay taxes. If we're losing people that pay taxes our social safety nets will crumble. Maybe that makes the math more in favor of actually using humans too.
I'm fully in favor of a social safety net (a proper UBI even), but the tax system should tax what we don't want / keep within bounds, and I think in general we want more automation and more free time. So we should tax resource consumption. Short time, this might turn out the same (i.e. whether to tax agents through electricity consumption or agents directly might not make much of a difference) but I think long term this would align incentives across a few dimensions very well.
It should be a clue that most of the people making the rules are generally considered nonproductive, or at least adjacent to and aspiring to be nonproductive.
And I think that is a major mistake, has many unwanted side effects, and only works because work supply does not yet exceed demand, so the taxes still keep flowing. But the relationship is beginning to break and I think a resource consumption based model helps with two very current challenges of society.
The people who make the rules (by brainwashing the population through a trillion dollar advertising industry what would make Pravda blush) only want to help themselves. As long as they have enough toys to prevent a French Revolution or for other people to take their wealth, they’re happy.
200 years ago, even 1000 years ago, the rich had to live in the same towns as the masses so they did actually put effort into keeping them nice. Building amenities, providing jobs, etc.
They don’t have to do that now. Power extends far beyond a 20 mile radius. They don’t have to go anywhere near the people.
Taxing resource consumption, would likely encourage companies to optimize for less and less resource consumption. That by itself might be a good thing. But what if some day companies achieved the ultimate savings, using very little resource to produce the max amount of things. You'd get very little tax revenue to feed the society.
Of course, if those max amounts of things have few consumers because most people are just too poor to afford them, that'd make companies (or the agents that run companies) think twice too. At some point these AI agents might work together to come up with an economic system to make it possible for society to afford their products so they could further grow.
I think a resource consumption based tax model extends very neatly to a UBI model. One could think of as: "this is the basic resource consumption we allow each person" and it automatically generates demand for basic services everyone should be able to afford with their share of resources, while keeping a free market and allowing an economy on top with extra resources. Standard money could still be the means for value exchange, just taxing works differently.
Here's a wild idea: about the CEO and shareholders being properly taxed, instead? Every job being replaced by AI is with the purpose of making more money for the CEO and shareholders, right? So, tax them more. It would solve SO MANY issues
Where do you draw the line to define an "agent" that should pay taxes? Many jobs once done entirely by humans have been replaced by software or mechanical automation long before AI came along. The issue of eroding social safety nets due to declining income tax revenue can be solved in a much more straightforward manner that doesn't require complex AI tax schemes.
I don't think taxes are the main issue, but people losing their jobs.
We need UBI. Taxes are still there anyway, company income/profit tax, VAT, and maybe we can have other taxes too for automatically-generate income, same as with investments/dividend/gambling tax and stuff.
Where does it end though? The most optimistic (or pessimistic, depending on your viewpoint) say that AI will be able to replace or displace lots of professions. Is there enough work for everyone? Is there enough meaningful work for everyone? A software engineer or accountant making ~150k a year will find another job that pays less, so he will also pay less taxes. What's going to happen in the short term, inbetween this shift? A bunch of unhappy unemployed people leaning on social systems. That seems pretty devastating. Who is going to come out as winners?
Jeremy Irons's CEO character in _Margin Call_ explains his job like this:
> Do you care to know why I'm in this chair with you all? I mean, why I earn the big bucks?... I'm here for one reason and one reason alone. I'm here to guess what the music might do a week, a month, a year from now. That's it. Nothing more. And standing here tonight, I'm afraid that I don't hear - a - thing. Just... silence.
I'm sure many CEOs are bozos who can be replaced by a statistical averaging and guessing machine. But this skill of saying "it's time for us to dramatically change course because the music has stopped" seems to me very contrary to the current LLM paradigm. RL makes the chatbot agreeable ('aligned'). But changing course is inherently disagreeable. It makes enemies and people are going to try to talk you out of it at every turn. I think we'll need a new idea of what LLMs are for, and how to train them, before they can do that. (Much easier to mimic and therefore replace a Ballmer than a Nadella, in other words.)
Organizations/Biz requirements (be it scope, money, HR, etc) changes depending on the size thereof.
This assumes that whatever your biz is, it's at Google level. It's too much. Sometimes you need just few employees (few agents) to get the things done.
This app will overcomplicate flows just like in real life.
The only lesson I will take from app is how SenteLab created the service, and why (<- so you know when to use this or not)
I had this hypothesis back in 2014 that all the communications from higher level managers could be replaced by a markov chain, because it was just variations on the same themes ("improve conditions", "make it easier to do....", "<generic positive words that don't translate to actual actions or measurable improvements>".
I have a friend who works a high-level management position, who told me that as long as results are delivered, management is doing their jobs 'by definition'. So if a manager is either horribly abusive or totally incompetent and his subordinates make up for it in some way, that the result is at least acceptable, nothing bad can happen to said manager, at least officially.
Only failure is actionable, in which case said managers will make a good effort to shift blame.
So if your boss is an idiot and an a-hole, and you have to work twice as hard because of it, you are enabling the bad working conditions he's subjecting you to.
Thankfully my boss is actually great (he doesn't know me on here), and even the suits know enough to just leave us to get on with it while they play musical chairs.
Yank on that thread harder, developers and engineers are not safe just because they rely on different patterns.
But once you've toppled the house of knowledge work, you'll find you can rebuild it easily because the foundation still exists, and it's the thing missing in this analysis; judgement.
Servant leadership is a management cliche that works well as a counterbalance to the dominance hierarchy that's implicit in a human org chart. I'm not going to listen to a Claude instance telling me to do something I'd prefer not to do.
Your comment is entirely emotionally driven. “These politicians! These bureaucrats!”.
Let me guess, it just happens that the role that you work in is amongst the only that actually provide value, and everything else is just waste.
Developers are among some of the least self-aware people ever, I swear. And I say this as a developer. The God complex instilled in techies in the ‘00s and ‘10s is unrivalled and I’m really enjoying seeing some of you being taken down a peg.
They’re just doing what they’re told to do by the shareholders. They’re the largest cogs sure, but cogs nonetheless, and they’re replaced just as quickly when they fail to make choices that maximize profit above all else.
You guys are so daft about it. The fact that you do not see what work goes to keep company running and be responsible for thousands of people still having their jobs…
Some work is 24/7 basically, when you have to keep business afloat there are no sick days or worse days. Your brain is 24/7 preoccupied with the work.
Meanwhile you walk out after 8 hours and stop caring at all because - someone else does
Someone else grinds to provide you with employment so there is work at all and the company keeps sailing.
I am sure you guys are smart, adults. stop being so childish and immature. You are better than that
(Corroboration, not deduction: "has four legs, so it is more probably a cat". That is vs falsification: "short, hence not all managers are tall". Which does not work anyway going from the logical context to the stochastic context, because it becomes "the exception that confirms the rule". I trust you knew, but.)
> One of the miners, rejected on account of his size, offered to fight any man there as proof of his suitability as a soldier, and six men were eventually called upon to remove him.
Depends on the board. The premium CEO robot will be powered by a bio-ethanol fuel cell so that it can play golf with the board of directors for as long as drinks are available.
As I always say, CEOs are corporate status symbols. "Look how successful we are, our CEO is super-handsome, went to William and Mary, and we pay him $15 million a year"
Last 5-6 years: COVID surge in HN brings in more people
Last 3-4 years: General rise in AI topics which have broad appeal and brought in a lot of people
Last 2-3 years: More overtly political discourse on HN bringing in more people
Last 1-2 years: Bots start coming in, some are very obvious, but they have gotten better and better to the point its hard to tell anymore.
Last year: Sharp rise in shallow redditor-style comments, especially in cultural/political discussions comments that are snarky but contribute little to discussion outside of affirmations like "Laws are for poor people"
That’s how leaders throughout history have worked. Usually you prove your mettle in the first half of your career and then do the hard work of being the face of the company once you’re an executive. Diminishing what a leader does is the foremost underling cliché throughout all of history.
You're hearing the echo and ascribing it to the noise.
A constant push and pull between leaders and the led is the fundamental. Criticism in both directions is necessary and healthy, and at different times both correct and misplaced.
There have been great leaders and awful ones. Acting as if all leaders are proven and deserving is as asinine as naming all CEOs as useless.
Right now we are at a point where CEOs are biased towards being quite anti worker and too easily rewarded. It's not surprising the cliches against them come out.
I would argue the bias is for ceos (of public companies) to act a certain way due to inherent incentives the way the world is set up, which biases their personality. But I won't go into that old argument again.
I'm sure their AI can simulate bad executives very effectively. Good ones, not so much. Humorously, the set of agents don't include an AI CEO, just the execs underneath the CEO. Someone must set the goals, after all and current LLM's have no agency (yet... thank the gods). Finally, the tasks their LLM agents can supposedly do are more the tasks that the executive's staff handles, not the executive themselves. Unless it's a smaller company, I guess.
I have no experience with data on bad and good CEOs depending on their working context.
At this stage, this seems to be a demo, albeit an interesting one. I find the concept still very intriguing, as it is exactly the thing that attacks white collar functions in administrative and executive tasks sets. I need to give it a spin.
“Walking away costs you the deal. Signing costs you the company. The math is obvious in retrospect; harder in the moment when there's quota pressure”
such valuable, should I say load-bearing insights
Other gem from my new CFO:
“The financial statements are the result. The variance analysis is where the operating insight lives.”
And my new C(3-)PO just told me
“If we removed all sales involvement, what % of revenue would we still get?
#5 is the honest PLG test. If the answer is <30%, PLG is supplementary at best. If the answer is >60%, PLG is the actual motion and the rest is overlay.”
Searching for PLG, I found: “In short, companies that use PLG models gain another revenue-generating channel: the company’s product itself.” What a world we live in, that a company's products being a revenue-generating channel is an innovation worthy of note!
The only thing worse than a human CEO is an AI CEO with power to fire you. People just don't think it will do stupid things, have nonsensicle orders which they have to follow, and likely fire people left right and center.
But oh ho there is no human ceo, must be a good place to work!!
As a shareholder I’m optimistic about the possibility of using LLMs for senior management positions, and on down. C level executives are rare, expensive and often unpredictable quantities. LLMs, if proven to work, could be easily cloned, relatively cheap to run and as a bonus they’re unlikely to get involved in improprieties with their subordinates.
Why should this be an installed program that uses agent SDK API calls instead of a prose skill to be used with existing agents? It builds "eight specialist AI agents". I'm just wondering what kinds of requirements might cause you to decide to build your own agent architecture rather than use existing ones?
I think this is an interesting idea but the non tech parts need fleshing out a bit if only to fire the imagination
The way the title sounds is that they got fired, then they created a new CEO and continued as normal
In reality payroll would be gone, any customer relationships would be with the previous entity would be gone, any capital/fund raising whatever not helpful any more
I'm actually surprised we don't hear more about companies created where AI is the leader - I think over the next year or so when the commercial labs release persistent agents this will be more viable because right now its still very difficult to take seriously due to deficiencies in the models
Yeah, this is actually one of the best uses for AI, I think. Boring decisions that you have to take, but without all necessary knowledge, based on history that can be learned and summarised. And it's all a lot of luck, and nothing is 100% a good or a bad decision. Algorithms won't help. It makes total sense to me to make it openly heuristic and based on propabilities and therefore use AI.
Just like in Iain Banks's scifi (although those AIs are actually intelligent -- I don't think you need that for a CEO).
At some point someone will make a big company from zero managed completely by AI and the owner will just do investing. We just need an open source IA that get the IA hiring stuff.
This reminds me of William Gibson's "Burning Chrome", where the head of a syndicate is a Russian AI (Chrome) running things behind a smuggling front. Gibson basically predicted the AI CEO forty years early!
I think it's pretty fun. Worker cooperatives no longer need different ranks of executive officers. They can all be peers with different functions. So if worker cooperatives are going to succeed in the US, they have pretty good conditions right now.
Perhaps this AI CEO will bring an end of all this over-paying of top management. Running a corporation isn't rocket science; it's just a matter of putting profit above everything else. Any idiot can do that.
I've been slowly developing my hermes agent to do something similar but is very customized towards my needs. I should spin up another agent and give this one a shot and see what it's capable of. This is hilarious.
Highly configurable. Looks like it is targeted at any level of management. The key features seems that it can generate its own vision of how to move something forward and that it can be set with degrees of freedom depending on the use case. Video demo here: https://www.youtube.com/watch?v=O_g97xxVTMk
To the extent this is tongue in cheek it is funny, but I could imagine if I was building a startup I’d like something like this if it was good.
That said, if you look at what it’s actually doing (read the question carefully, then read what it actually said) I am not convinced this is better than what ChatGPT could do out of the box already.
I’ve been saying people need to do this for over a year. I’ve known (personally) quite a few executive leaders and while I don’t think AI is a good replacement for any _good_ person at a job, the executive team is the low hanging fruit.
We glamorize execs but at the end of the day if you watch their careers they have plenty of failures (usually) and generally fail upwards. I’ve known some amazing CTOs and VPs of engineering and they’ve brought more to the table but a lot of what the executive team does is looking at the currents in the industry and trying to make educated guesses about where things are going. Yes more than that but this is a news.y comment.
"CEO-abliterated" Model got uploaded to HF. Nvidia bought HF. U.S. State Department pauses immigrant visa applications. Developers went back home. CEO received bonus. 2.5x larger than last year.
Just like working with Ai on dev projects is like working along side a really smart but drunk buddy, an Ai CEO is like being guided by a coked up business coach. In the end, with Ai where it is, I think it could maintain a company already moving along with healthy systems and a known cycle. But same as where Ai fails on larger dev projects, the same happens with a business. The timing for stepping back and diving deep is critical to seeing opportunities for growth, avoiding pitfalls, and doing the daily maintenance required to keep any project alive in the long-term.
AI CEOs are the way of the future. If you want to reduce overhead costs, the maximum save will be produced replacing the bloated salaries of the C-suite. This is the most inevitable ROI play presented by AI.
I think AI CEOs are dumb but AI 'enterprise decision-makers' could be very very interesting.
Imagine how quickly your meetings could get resolved if you had teams submit prompts / contexts, had a clear, documented set of overarching objectives, etc.?
This was fun and it will probably produce some nice output, but developers in particular, have a tendency to don´t see the value in being able to inspire and lead people. Being a leader is mainly people work, and AI cannot replace that.
If human-led companies started consistently losing to AI led companies... would it say more about the intelligence of AI, or about how thoroughly power corrupts humans?
Many CEOs are bad and their businesses will either fail or be bailed out. But the work CEOs do is probably some of the last to get automated - higher order strategy, relationship building, managing people.
This is good for satire, but the real solution is to use AI to become the CEO, or take over the scope of of product or decision-making roles. There are many business-savvy technical workers that get pigeonholed because they spend most of their cognitive effort on low-level technical details. This is an opportunity for these people to rise up and prove they work better than people who manage or direct them. CEO is really not a role to automate, unless it's a fully autonomous business that has zero humans.
If people would unironically start pushing for this, I genuinely do not know what to do.
I guess that would be at least a nice bang to let democracy finally die.
Let's hope that we never see that happening.
___
To expand on that:
I believe that these thoughts are so incredibly harmful to humanity itself, (and, more importantly, I believe that the people in power believe that too,) that we'd be seeing pushback against this.
Maximum pushback, because what the pushback would be defending would be - as said - humanity itself.
Said pushback would be bypassing democracy, because there is no way of arguing with these crazy thoughts. Which would said a precedent, because why stop there.
Which, again, makes me believe that this is some sort of psyop, because I refuse to think that the people on here would really be as blind to the systemic implications of their brainfarts.
___
FWIW, a ruling class worth their net worth would go for the less draconian thing of simply taking away your internet access and forcing you to touch grass.
One that is.. uh.. less signed up on the whole "all people are equal and worth saving" would line up.. idk.. 100 people with these horribly bad-for-humanity ideas and shoot them in the head live on stream or something.
The interesting twist here is that LLM rulers might be more biased towards doing option B, as the statistical probabilities of words cannot _feel_ the chemical hormone cocktail invoked by the mental image of that.
One thing AI is actually good at: look at a bunch of data, make informed, unbiased, objective and quick decision, have a backup plan if that fails.
This is exactly what an executive is supposed to do. But since executives are the ones in charge, nobody talks about this, it’s all about getting rid of developers and how engineers arnt needed :).
Oh, not in that sense. I meant if you showed it data and business metrics, and asked for s tough decision, it is likely not going to make a decision by factoring in the race of the manager whose team isnt doing great, or get influenced because another executive is its beer buddy, or because the finance guy used to work with the AI at a previous job.
I am talking about these biased a human manager or executive will most certainly have. We have all seen it.
A team of software developers were recently let go so a company could replace them with AI. So they got together and created an Open Source AI CEO to replace the CEO and other executives. Hopefully, turnabout is fair play and this might even get some folks to think twice about using AI to replace people.
Corporate officers, including CEOs, have to be "natural persons" (legal jargon for flesh-and-blood humans) either directly by law or indirectly as a practical matter so the attempted turnabout falls flat on its face.
Sun Tzu said to "know your enemy" and those who don't know the basics of how businesses and finance work have no hope of defeating them.
I’d be more interested to see a collection of independent guild, unions, and co-ops that voluntarily coordinate through an AI intermediary.
Because the AI can’t really own things or be held responsible, the idea of a skeuomorphic AI seems quite silly. Keep the ownership in the humans, but plural, and have them in contact with the actual work.
This is one of those rejoinders that tries to be clever but really, really isn't. The "intern as fake CEO" is left holding the bag because it's their signature on contracts and other legal documents. The idea of puppet CEOs to fraudulently dodge accountability existed well before LLMs became a thing. Courts see right through that kind of scheme plus the "intern" is likely to turn coat against the mastermind to save their own skin at the first sign of trouble.
If you think CEOs are the primary legal signatories for their company you might need to spend some time away from garage band startups. Companies maintain entire legal departments for a reason.
Maybe, maybe not. A lot of folks would gleefully avoid rocking the boat if handed a job this cushy. If they do flake out you could ask if anyone in facilities would like to take a break from emptying trashcans...
This is hilarious, I feel like we are going to see AI companies add something to the system prompt or terms of service that explicitly forbids having an LLM act as CEO. Honestly though I do hope this gives pause to the C suite when it comes to replacing people with AI
I dislike performative business people as much as the next guy, but you can't tell me that you people really think that this is anything other than a petty and equally performative PR stunt.
If you want to see that idea extrapolated to an impressive extent, check out Accelerando by Charles Stross. If Stross was right we're in for a bumpy ride.
I'm really not sure why people are cheering for this.
We've been rightfully pissed at companies doing unethical things and getting away with it, as there is simply nothing you can hurt them with.
Why is HN fantasizing about making the companies even less hurtable? Are people here insane? Otoh, a lot of those accounts are just a few months old, so.. welp. Nice psyop we got here.
We're all joking around here. But 5 decades down the line, with increasing cognitive offload to AI, is a slow erosion leading to this end state so crazy?
Everyone seems to be missing the point of the role of the CEO.
The CEO is not some masterful person answering all the questions for the company, I would argue that a CEO should be able to disappear for a year and if they had done their job correctly and assembled a good team the company would do just fine without them.
No the purpose of the CEO is to safeguard the Capital that has been invested in the company. That is all. They make sure it doesn't walk our the door in wages or other wasteful spending. They are the Capital's man on the ground.
Yes, this is actually a reason why an AI executive could be a formidable alternative in theory.
The CEO is ultimately an employee of the board, and managing the CEO with respect to the board's oversight and interest is a somewhat adversarial challenge.
However, the AI CEO's intentions are fully transparent to the board and entirely pliable to the board's demands. The AI CEO also offers the board the opportunity to retain significant equity shares that materially impact their wealth - in this way AI executives stand to be an even more tantalizing target than fixed labor costs.
Of course, this presumes LLMs will become reliable enough for this, which doesn't seem likely.
I would argue that many companies would in fact be better off disappearing for a year, rather than constantly thrashing the company to chase the latest trend.
And judging from the ZIRP hiring sprees and tokenmaxxing, CEOs are more often the driving force behind wasteful spending than they are the safeguards against it...
If a ceo can disappear for months and still return into a company that isn't on fire,it means they have a functional company. It means there are processes and people in right places to run things in a continuous manner. This isn't as common as people might think. I've seen many companies where people can't survive for a few days without ceo because the position is tied to a million things( often unnecessarily). There's an old saying about owning a business: if you leave and it still runs it means you have a business, otherwise it's just a job
I’m all for productive people, brilliant engineers. I still get dirty with codes/work all the way down to the newly hired fresh out of college kids, to the production architects that shapes multi-million, if not billions, of P&L. I’m also very lucky to be included in key decision making, planning, negating amongst the boards of C-Suites, Directors, VPs.
A very personal view but many developers/engineers fail to see the overall picture. And yes, I was there and I went through all of the mistakes.
The things you see on the surface of those incompetent CxOs are just the tiny piece of the bigger picture. You cannot replace blames, decision in the face of uncertainty. Almost all decisions that businesses take have nothing to do with the proficiencies of technologies or engineering finesse, or even the best options, but mostly on “incentives” and who can take/own the “blame.”
“AI” replacing developers is evident because that is a rather tiny perspective and kinda easy to do the equation - developer can do that (good, bad, best) and can the AI do it - maybe, perhaps. So, replace.
A C-Suite took a decision to tick the final approval of a billion dollar deal with a service providers (amongst 10s of them). The idea might have sounded idiotic because there were 3 others better at the top than the chosen one. Even if all of the task and options were weighted with AI, but that last human decision is the Blame that the C-Suite took on, by betting his incentives on it (money, fame, pride, ego.)
I think a lot of this kind of talk treats the C-Suite as special, mystical "visionary" people who Know Things We Can't Know, and by virtue of this otherworldly magical power, are uniquely qualified to do their jobs. Like the average person with basic knowledge of his business and industry can't possibly choose one service provider out of 10 bids, or set business priorities, or drive and enforce culture. We're not talking about wizardry here.
People always dunk on me for saying this outrageous statement, but if you put me in charge of a random Fortune 500 company, I give myself a 50% probability of the company performing better than if its real CEO was in charge, and a 50% probability of it performing worse. I really don't think these guys bring as much to the table as they think they do and as much as the general public seems to think they do. Business success does not come from paranormal insight and abilities.
I think you could randomly nominate CEOs from the set of the general population with, say, a minimum of a college degree, and the randomly nominated CEO's performance would not statistically differ from "qualified" CEOs.
There are some exceptional executives out there with strong vision, ability to execute and build. Those are rare. The rest are just a bunch of people who had risen to the position because of whatever. They often don't have any special knowledge or skills and succeed mostly because of how the system works. For instance, I can't imagine what one would need to do to fuck up in a company like Coca-Cola. It's super well established, supply chains are golden, everything works,etc. So does success in such a company really mean the executive is some sort of visionary genius? Absolutely no.
CEOs are probably by and large less technically competent than the average technical line employee, but what companies are actually paying for is personal connections to power structures that can be leveraged for profit. Look at someone like Condoleezza Rice- almost certainly her political science doctorate imparted her with no real technical knowledge, yet she's served on the boards of C3.ai, Chevron, HP, Dropbox, Transamerica, Charles Schwab and a million other vague leadership positions for other entities in fields she almost certainly lacks any competence in.
This is absolutely spot on. CEO/C-Suite/VPs are surrounded by gatekeepers (usually something like the PMO org, program management) who feed them with garbage info. Even if a top level leader is genuinely interested in making things better, they would never know because the gatekeepers firewall any competent person.
The gatekeeping org has the highest density of incompetence. They only hire incompetent because loyalty to the person hired is the most important. Gatekeeping orgs of various leaders protect each other, there is almost never any blame, it is things are good and they are going to get better, announce xyz initiatives/projects.
> Almost all decisions that businesses take have nothing to do with the proficiencies of technologies or engineering finesse, or even the best options, but mostly on “incentives” and who can take/own the “blame.”
This argument always confuses me because it seems like lower level employees are the ones that get quickly dismissed when things get bad while a lot of C level executives seem to have some sort of compensation even if they fuck up.
> A very personal view but many developers/engineers fail to see the overall picture. And yes, I was there and I went through all of the mistakes.
What is failling to see the overall picture? I hate this overall picture mysticism. I bet that if people where given the same role and information as the CEO they would be able to see the overall picture.
If a company embraces AI at all, this should be the first step. Replace the MBA bots with actual bots. Imagine the amounts of budget becoming available for sensible purposes.
Not every company needs a CEO, and that is ok. Companies do however need a decision making process and clear authorizations.
Creating an AI executive suite can bring some skills to the table, but it doesn't solve the authority question. I'm curious how this would work in practice...
What happens if the AI decides it wants to commit funds - does it have authority?
What if it wants to change direction? What if it wants to fire a developer?
I really want to see a proper ai team that successfully replaced a bunch of people. I have not seen it yet. Only thing I hear is gold plated stories about the potential of agents running fully automated workflows without people touching a single lever.
In the meantime I am using AI almost the entire day. I have to discuss these topics with managers that only use chatgpt and watch youtube videos from people disguised to promote their own crap AI products. All I can say is that AI is not good enough to replace entire teams. It can do some cool stuff. But automate entire workflows. Nah. It's just not gonna work.
Op here. Team was let go. Decided to do a startup. We thought that tools like co-pilot and Claude code were transforming development. We wanted to transform management (not just CEOs). We open sourced this knowing that if people find this useful (and we hope they do) we can get a support work out of it (similar to the Redhat model and other open source tech).
The difference is that the CEO controls the capital and has the ear of the investors, whereas the developers generally don't. It's the investors that control whether the CEO is a bot or not.
Number of people who think that a leader or a CEO can be replaced by AI and who themselves have actually risked everything to start a company, be a leader, or do anything similar: exactly 0.
Interesting point, but CEO is not necessarily the same as owner.
For those that missed it, the question what current owner would turn over control to an AI?
I could more easily see a individuals delegating work to an AI but retaining actual approval/veto power. In that sense, the approver is basically still the CEO .
> Number of people who think that a leader or a CEO can be replaced by AI and who themselves have actually risked everything to start a company, be a leader, or do anything similar: exactly 0.
Prove it.
> be a leader
There are a lot of bad "leaders" out there running Fortune 500s, fwiw.
If you just want a sociopath with faux accountability that's not very difficult, dime a dozen.
The world is a whole lot more complicated than the simplistic "CEOs are evil" nonsense. I suggest you talk to an entrepreneur sometime. Or, better yet, do it yourself before judging others so harshly. Or is it too hard for you?
> I suggest you talk to an entrepreneur sometime. Or, better yet, do it yourself before judging others so harshly.
I've been a founder, CEO, and combat leader under fire in Iraq.
> Or is it too hard for you?
I guess, clearly, it wasn't. Have you ever lead men in combat when death is real and you are right there as the tip of the spear, with some of the smartest and most high level athletes in the world?
A lot of fortune 500 CEOs are nepo or network hires to grease wheels and laws, but even then, it's not necessary merit or performance based. There are some good ones, but there are more bad ones than good. All will write books on "leadership"!
A lot of you calling things "leadership" in large companies can be done with fairly basic calculations and you don't even need a high functioning sociopath to take the blame. That doesn't mean you don't have a board judging performance.
Also a lot of you are taking things very personal, thinking most of these comments and complaints are talking about some CEO of a 5 or 500 person company, but what many of us are talking about are companies with 50k-100k employees or more (then there's Amazon with over a million).
But yeah...
> "Or, better yet, do it yourself before judging others so harshly”
Honestly the best use of AI at the moment is to replace non-technical middle management layers. Many of those yelling the loudest for “more AI now” are quite ironically holding the jobs best replaced by the current creation of AI.
Great, now this open source CEO can fire people automatically, and replace them with open source agents. The whole organization finally gets replaced with AI. Time to invest and become a shareholder!
This has likely been the case as long as jobs have existed. Some CEOs spend their entire time tweeting, apparently.
The way any large company works, and this includes all the big tech companies, is that they end up in a state of constant churn. Every 6 months or so there's a reorganization. There'san old joke that you get an email that your manager's manager's manager's manager now reports to a new manager and you've never heard of any of these people.
The point of that churn is to make sure nobody is ever held responsible for something not working. It's never given enough time. That's by design. Middle (and even senior) management is about empire-building.
Another quote that springs to mind is that the bureaucracy is expanding to meet the needs of the expanding bureaucracy. That's how I feel about management bloat.
Anyone who has paid attention to Intel's downfall in the last decade will have heard stories about how the company descended into fiefdoms and it became impossible to get anything done. That's what happens.
Most people in these organizations aren't particularly talented at anything other than forming social relationships. the leadership of these companies always end up having an adversarial relationship with their workers and try to suppress their wages to increase profits so they can get their stock options and bonuses and the CEOs all sit on each other's boards of directors setting each other's compensation packages. It's a lootocracy.
The point of all this is that the leadership will never decide that they are the problem. Take Meta's Metaverse. It was a $70 billion flop. A lot of people got laid off from that. Were any of those worker bees responsible? Nope. They didn't choose the strategy or probably even what they worked on. It takes a lot for heads to roll within senior leadership and with the vote-stacked share structures of many major tech companies, ousting a CEO is essentially impossible.
These executives think shifting peas around the plate is paramount to eating them. That's what toddlers do when they want to try and make mommy think they've done a good thing.
No one wants to just get down to the business of choking down these damn peas.
ChatGPT (or something from OpenRouter) should be the default model. Stop promoting Anthropic's business when they're so hostile to the developer community.
Power is what makes them what they are, not knowledge or skills or empathy.
If you don't like what this thing tells you to do, you just won't. It's not going to fire you. Therefore it's not going to steer the company away from disaster, or towards something good. It will be just a talking box.
> Power is what makes them what they are, not knowledge or skills or empathy.
No. Oh my god no.
This is just engineers coping with the reality that they've been filtered by skillchecks of a system they do not (yet) understand. But instead of doing the engineering thing and starting to understand said system, fox and grapes gets deployed.
Which isn't to say that there would be zero examples of what you describe, but the jump to "all are like this" is crazy, and, more importantly, anti-intellectual.
We've replaced our CEO for the better part of 6 weeks now. He has been on a media tour promoting a product we no longer sell, but it keeps him busy while we continue to grow.
While Proxy AI [0] is not open source, it is free, it is built from pure garage energy, and we are having our summer sale right now. 20% off for HN users.
reply