50% of what middle management is what secretaries used to do. If I ever run my own company I will hire a team of secretaries to work with my developers and our productivity will shoot through the roof. No everyone is organized, good at scheduling, following up on things, focusing at work, interpersonal discussions, sending memos, navigating bureaucracy, etc.
A great secretary allows you to focus on getting development or whatever work you do in and out instead of on the day to day running of life.
I don't want a scrum master, I want a secretary.
As a manager, 90% of what I do as a manager is follow up with people, act as an escalation point for managing timelines, organizing who can do what, and none of this really is a skill related to my actual skillset.
Luckily I am able to commit to 20/hrs a week of development, but I could probably do 35 with a secretary - and use those last 5 hours to focus on strategy, direction, leadership stuff that's actually important.
Hallelujah! I felt this exactly when I was a director managing about 40 people. I would say at least half my day was taken up with logistics. The amount of actual mentorship and engineering leadership I did was less than the amount of time I had do to compared with a lot of bureaucratic responsibilities of my job.
At the time, there was an admin assistant who worked for our CTO, and she was fantastic - extremely detail oriented, organized and proactive. I kept thinking "All my training is in software engineering, and yet I'm doing all these tasks that she is much better at." Kept thinking that if my job was "split" down the middle that everyone would be much happier and my company would be getting a much better deal, with people's time largely spent on the stuff they are best at.
So much this. And also this applies to many roles. I heard similar from people who ran buying and merchandising teams at my old company: “Why are they paying me to file my expense report? An assistant could do this and then I could spend an additional hour doing my job!!”
I hadn’t ever really thought about this until then. It seems like most companies just think “having an assistant or secretary is extravagance” so obviously they can’t afford it. But if they really knew what kind of unskilled grunt work highly-paid people spend so much time doing I think they would change their minds.
Completely agreed. I was a marketing director for about a year and a half, and I stepped back from it because I was personally spending about 60 hours a week on items I would consider management, which I've never been that interested in. I'm interested in leadership; in building productive relationships; and excelling as both a marketer and developer.
If I'd had a secretary or assistant or something to handle keeping on top of projects and KPIs, I would have been able to focus on being great as a leader. Alas, that wasn't an option, so I got out of the way for someone who excels as a manager to take my place.
On the plus side (and admittedly an aside as well), stepping out of that role allowed me to write and publish a book in 6 months, which is something I'd always wanted to do but couldn't find the time for.
People forget the plan recommended by Brooks in the Mythical Man Month. Surgical teams with delegated duties led by an experienced surgeon. Positions of a secretary, language lawyer, documentation writer, (and others I forget). I would love to just once work or lead such team.
Whatever happened to secretaries anyway? Were they really cost-cut for managers to do by themselves?
You're right though, it's a valuable job being the glue of a business. In more than one way, not just work-wise. Social stuff matters too, and was often the domain of the secretaries.
I heard a similar sentiment from a high performing colleague from a company my employer partered with. He said something along the lines of "if they (client) wants me to make constant PowerPoint reports, they better pay extra for an intern in the contract because I am not wasting time on that".
I have almost this exact conversation with folks all the time. The majority of managers are dead weight and often have a negative impact on performance and culture.
Leaders (natural or learned) are incredibly valuable.
This already happened once, in the 80s. A massive reduction of the 'middle class', by eliminating massive amounts of middle management.
Managers aren't overhead, it's an absurd statement, because it's an absolute. Managers allow leaders to extend their reach. Absolutely no leader can scale beyond a certain point. At all.
Managers allow for that. They're essential, beyond a certain org size.
Now, I will agree that some should not be managers! But that is the same for any profession. You need to churn, to chop off those not performing, or poorly suited to the task.
AI could replace bad managers, of which there are too many, but not good ones. Can an AI provide a sense of connection to higher levels or other parts of the company? Not even possible. Can an AI motivate? Can it resolve conflict? Doubtful at best. I certainly wouldn't trust one with any of the more therapy-adjacent things that a good manager does, and that's a big part of the job if they're doing it right.
The problem with so much management - the reason the rest of us hate it - is that it's already too formulaic, too inhumane and empathy-free, too much of people as machines rather than as people. Replacing that with actual machines is not the solution. It's just more cost-cutting at the expense of workers.
(Yes, I realize OP was satire. The points on both sides still stand.)
When the headlines were “AI is going to cause a loss of tech jobs“ I heard “AI is going to manage my company (the part I don’t want to do) while I build the things that solve customer problems (the part I do want to do)”
As an engineer for 40 years, I don't need to be led. I know what needs to be done.
I would say "leadship" is a highly emotional (on both sides) behaviour that would be perfectly fit by a solution based on historical statistical analysis.
I'd say there are most CEO's that probably could be replaced, assuming that CEO isn't also the founder, you can't replace the 'visionary leader' of a platform, or if the CEO is someone like Steve Jobs who I guess was sort of a founder of apple so that's kinda moot.
80% of CEO's could easily be replaced in a few years, then that can trickle down to other managers, and eventually lead towards more "flat" organizational hierachies which hopefully are more equitable and pay better for all employees, not just a rich few.
Rao's The Gervais Principle would argue that middle managers are essential.
His interpretation of them is that they act as the carbon control rods in the nuclear reaction of capitalism. Prior to their invention, we had workers and owners. And then we had a lot of strikes and worker solidarity and other nasty socialist things like that. Then we invented middle managers for all the workers to get mad at instead and shield the owners from all the ire and rock throwing.
Rao's interpretation is overtly tongue-in-cheek, but makes a good point. That middle managers aren't strictly there for productivity, but act also in an emotional/human-nature/psychological capacity.
This isn't a vague term, it means people who manage managers, but who aren't the very top level. You'd have to be pretty delusional to be actively managing a group of managers/teams and not realize you are middle management.
Which is not to say that AI won't be immensely helpful for running companies; just that once you're in the role for two days you realize that replacing that role entirely isn't going to happen.
> AI can win at chess now. You think being a CEO is harder than chess? Come on.
Being a CEO isn't like playing chess. Being a CEO is more like a really complex version of self-driving cars. Why? Because it involves interacting with other humans, all the time. It's all people problems.
Chess isn't. Chess is decision-making with a finite (if immense) set of possibilities. Very different set of problems.
The kind of journalism that journalists want you to think they do - going out into war zones or directly interviewing people to find out new things of public interest - isn't replaceable by AI. The kind of journalism they actually do - hanging out on Twitter and repeating whatever goes trending there - is easily replaceable. They did this to themselves.
Chess also has a rigid, well-defined set of rules. Business does not. Being a CEO requires combining a lot of (sometimes nebulous) mental models, each with a degree of uncertainty, to some overall business strategy. I think it was Sam Harris who pointed out this is why there are child prodigies in subjects like math and chess, but none in something like political science.
> mental models, each with a degree of uncertainty, to some overall business strategy
That actually sounds a lot like model based reinforcement learning. AlphaGo still had an explicit model of a perfect information game, but newer algorithms are capable of learning the rules of games like Minecraft with very little prior information.
The problem is that the real world is still much more complicated and hard to access for machines. However, I wouldn't say it is a fundamentally different problem, but rather a matter of scale. That's part of the reason why AI research spent so much effort on building game playing systems.
If CEOs were as replaceable and disconnected from the outcome of a business as the author suggests, boards would've started doing this long ago and CEO salaries would come down as a result (lots of people want to be CEOs, and by this author's logic, virtually anyone can do it). This is so disconnected from reality it reads like satire.
They replace C-level people often and the relationship isn't what you imply.
There's a structural power relationship in play. Monarchs play a fairly unimportant role these days but many countries still have them and compensate them handsomely.
CEOs are structurally capitalist monarchs. Sometimes it's even a hereditary position. They even do the corporate equivalent of court and heraldry stuff.
They're compensated not based on competency but instead, like every other job, via social arrangement.
If you don't think that's how salaries work, you're getting underpaid.
It may look like that, but it's not true. You know how a good system administrator doesn't seem that important because he prevents all the fires that a bad one would heroically put out? The mere presence of a monarch, not even his actions, acts in a similar fashion. They don't have to actively govern the country, but they have emergency powers that would allow them to prevent a wanna be populist dictator. And because of obvious game theory implications, these powers never have to be used — their mere existence, and everybody bring aware of their existence, is enough for deterrence.
Would they? I understand that in many cases, CEOs are often selected by, and their remuneration awarded by, boards made up of incestuous groupings of people who rely on each other in the same way. Their incentive is to award big paydays to each other.
Do you have any evidence that boards would not do this?
> I understand that in many cases, CEOs are often selected by, and their remuneration awarded by, boards made up of incestuous groupings of people who rely on each other in the same way.
Lots of people want to be CEO (until they become CEO and until they make CEO kind of money)
By the first 3 months reality sets in and the person now in the CEO role understands that they are so far removed from the action that they are essentially relegated to the role of spokeperson and cheerleader, at maximum the role of general moving imaginary troops on a fictional battlefield.
If they can make peace with their role of political spokeperson and cheerleader they'd bail out as soon as they reach 5M net worth no debt. No less than 95% would bail out.
The remaining 5% would bail out once they reach 10M net worth no debt.
CEOs turnover and population is so low because it takes a special kind of monodimensional individual to have the world as your oyster 5-10M in your name and no debt and turn around and say "I'd rather go to work and pretend to be at the helm of an imaginary ship instead"
> .. boards would've started doing this long ago ..
You are assuming here that boards are always acting in the best interest of the share holders. I bet there are many who do, but there are at least as many that act primarily in the interest of themselves.
Board membership is a lucrative job without much supervision, if any at all. So nothing is keeping board members from taking it easy. Why rock the boat if this can make you known as difficult? You might not get that 2nd or 3rd board membership then anymore.
You might remember Credit Suisse, the major bank that had to be taken over recently by UBS to avoid bankruptcy: until then it had gone from scandal to scandal for 15 years in a row. 2 CEO's had been sent away in that period (by shareholders) but the president of the board? He could stay, and with him the rest of the board.
His salary? 7 million a year.
I mean i think you could probably argue the long tail of poor performing small simple companies that sell widgets you might be able to automate a chunk of tasks. Thats as far as I would be willing to go. Other than that its a dishonest or uneducated argument.
The real world does not work like that. If you studied at the right institution or your family knows the right people, you will get the right positions.
The whole purpose of title inflation is to direct money to your friends with plausible deniability.
Now someone will bring up the rare cases where a CEO actually did something. That does not disprove the rule.
> Shareholders are spending these extravagant sums for tasks that can now be done just as well by an AI program trained on 100 years of corporate reports and Wall Street Journal texts.
Citation needed, etc. I think this is just meant to be funny but if anyone wants it to be more than that I’d need to see some actual proof.
> Shareholders are spending these extravagant sums for tasks that can now be done just as well by an AI program trained on 100 years of corporate reports and Wall Street Journal texts.
The quote from the article is clearly from someone who has never run a company, of size at least.
People have a very hard time being good CEO's, AI will not make that happen any time soon.
It's good to be skeptical of this kind of claim of course, but part of the point of the article in my view is that the narrative is pretty credulous when you say this about writers or programmers or artists -- i.e. people who do not control capital.
And yet there's plenty of evidence that certain copywriting jobs can be replaced by GPT right now. I haven't seen any evidence that GPT is competent enough to replace a CEO position, nor am I sure who would be writing the prompts in that case and how that position would be any different than a CEO.
(I'd be cool with CEOs making less than 400x their workers though)
CEO's and mid to upper management are about power, not doing things. Why is it so hard for us to understand something that is so simple and so obvious?
Great, with AI management maybe asset owners will finally get a board of directors who represent their interests and not the interests of the board themselves.
Truly successful CEOs make decisions where there is no definitive data and end up being right. It's not clear to me AI's would be good at that. Most CEO's have a chief of staff, that's probably something AI could replace.
Funny thing being in a big company doing a job that's mostly pointless.
I'm paid well. Do everything I'm told. Top performer. I'm happy though as I'm providing for the family. My accomplishments amount to one day of effort, output, results every month.
This has been happening for years.
I am curious how many of me exist. The Twitter firing suggests that's it's over 50% so it's that's true, what exactly are managers doing?
Most tell me they're in meetings, looking over spreadsheets and meeting with staff. I run my house and meet with my spouse for about twenty minutes a quarter over financials. Everything is automated. Everything is scheduled. We are adults. Labor is divided, assigned, adjusted as we age.
Large companies get in the position of not reducing head count because, yes, they're people with lives and relationships matter.
Solution? I'm thinking more of a State style employment but limited. If hired, you're job is secure for five years. Automate you job entirely? Good on you! Do not apply...... TAKE a new job in the company. You're slow to onboard? Fine. Dead weight? Not renewed.
Take the fear of insecurity off the table. People will behave according to their character.
Readit News
All the things mentioned in the article really are prime candidates for automation. But that's management, not leadership.
Management is primed to be automated. See Marshall Brain's book "Manna" for a great take on what's possible here.
Leadership is not. LLMs are great at doing the right things when there's a general consensus on what the right things are. Best practices, etc.
But what the current technology cannot do, and what I think LLMs specifically could never do, is know when to break the rules.
As Phil Knight may or may not have said (he was quoted in the new Air movie): You're remembered for the rules you break. That's leadership.
A great CEO needs to be a leader first and foremost, and a manager second.
A great secretary allows you to focus on getting development or whatever work you do in and out instead of on the day to day running of life.
I don't want a scrum master, I want a secretary.
As a manager, 90% of what I do as a manager is follow up with people, act as an escalation point for managing timelines, organizing who can do what, and none of this really is a skill related to my actual skillset.
Luckily I am able to commit to 20/hrs a week of development, but I could probably do 35 with a secretary - and use those last 5 hours to focus on strategy, direction, leadership stuff that's actually important.
At the time, there was an admin assistant who worked for our CTO, and she was fantastic - extremely detail oriented, organized and proactive. I kept thinking "All my training is in software engineering, and yet I'm doing all these tasks that she is much better at." Kept thinking that if my job was "split" down the middle that everyone would be much happier and my company would be getting a much better deal, with people's time largely spent on the stuff they are best at.
I hadn’t ever really thought about this until then. It seems like most companies just think “having an assistant or secretary is extravagance” so obviously they can’t afford it. But if they really knew what kind of unskilled grunt work highly-paid people spend so much time doing I think they would change their minds.
If I'd had a secretary or assistant or something to handle keeping on top of projects and KPIs, I would have been able to focus on being great as a leader. Alas, that wasn't an option, so I got out of the way for someone who excels as a manager to take my place.
On the plus side (and admittedly an aside as well), stepping out of that role allowed me to write and publish a book in 6 months, which is something I'd always wanted to do but couldn't find the time for.
You're right though, it's a valuable job being the glue of a business. In more than one way, not just work-wise. Social stuff matters too, and was often the domain of the secretaries.
I have almost this exact conversation with folks all the time. The majority of managers are dead weight and often have a negative impact on performance and culture.
Leaders (natural or learned) are incredibly valuable.
Managers are overhead.
Managers aren't overhead, it's an absurd statement, because it's an absolute. Managers allow leaders to extend their reach. Absolutely no leader can scale beyond a certain point. At all.
Managers allow for that. They're essential, beyond a certain org size.
Now, I will agree that some should not be managers! But that is the same for any profession. You need to churn, to chop off those not performing, or poorly suited to the task.
The problem with so much management - the reason the rest of us hate it - is that it's already too formulaic, too inhumane and empathy-free, too much of people as machines rather than as people. Replacing that with actual machines is not the solution. It's just more cost-cutting at the expense of workers.
(Yes, I realize OP was satire. The points on both sides still stand.)
Deleted Comment
As an engineer for 40 years, I don't need to be led. I know what needs to be done.
I would say "leadship" is a highly emotional (on both sides) behaviour that would be perfectly fit by a solution based on historical statistical analysis.
80% of CEO's could easily be replaced in a few years, then that can trickle down to other managers, and eventually lead towards more "flat" organizational hierachies which hopefully are more equitable and pay better for all employees, not just a rich few.
His interpretation of them is that they act as the carbon control rods in the nuclear reaction of capitalism. Prior to their invention, we had workers and owners. And then we had a lot of strikes and worker solidarity and other nasty socialist things like that. Then we invented middle managers for all the workers to get mad at instead and shield the owners from all the ire and rock throwing.
Rao's interpretation is overtly tongue-in-cheek, but makes a good point. That middle managers aren't strictly there for productivity, but act also in an emotional/human-nature/psychological capacity.
I think most people associate the term middle management with the guy immediately above them.
Honestly, a lot of this resembles the "If only the Tsar knew!" mentality in old, imperial Russia.
Which is not to say that AI won't be immensely helpful for running companies; just that once you're in the role for two days you realize that replacing that role entirely isn't going to happen.
> AI can win at chess now. You think being a CEO is harder than chess? Come on.
Being a CEO isn't like playing chess. Being a CEO is more like a really complex version of self-driving cars. Why? Because it involves interacting with other humans, all the time. It's all people problems.
Chess isn't. Chess is decision-making with a finite (if immense) set of possibilities. Very different set of problems.
Journalists might very well say the same thing (when people who've never been journalists talk about replacing them with AI).
That actually sounds a lot like model based reinforcement learning. AlphaGo still had an explicit model of a perfect information game, but newer algorithms are capable of learning the rules of games like Minecraft with very little prior information.
The problem is that the real world is still much more complicated and hard to access for machines. However, I wouldn't say it is a fundamentally different problem, but rather a matter of scale. That's part of the reason why AI research spent so much effort on building game playing systems.
I believe every MBA Program, Management Department and Executive Coach would beg to differ
There's a structural power relationship in play. Monarchs play a fairly unimportant role these days but many countries still have them and compensate them handsomely.
CEOs are structurally capitalist monarchs. Sometimes it's even a hereditary position. They even do the corporate equivalent of court and heraldry stuff.
They're compensated not based on competency but instead, like every other job, via social arrangement.
If you don't think that's how salaries work, you're getting underpaid.
Court politics and C level politics are functionally very similar too.
It may look like that, but it's not true. You know how a good system administrator doesn't seem that important because he prevents all the fires that a bad one would heroically put out? The mere presence of a monarch, not even his actions, acts in a similar fashion. They don't have to actively govern the country, but they have emergency powers that would allow them to prevent a wanna be populist dictator. And because of obvious game theory implications, these powers never have to be used — their mere existence, and everybody bring aware of their existence, is enough for deterrence.
Do you have any evidence that boards would not do this?
Can you give one or two examples?
Lots of people want to be CEO (until they become CEO and until they make CEO kind of money)
By the first 3 months reality sets in and the person now in the CEO role understands that they are so far removed from the action that they are essentially relegated to the role of spokeperson and cheerleader, at maximum the role of general moving imaginary troops on a fictional battlefield.
If they can make peace with their role of political spokeperson and cheerleader they'd bail out as soon as they reach 5M net worth no debt. No less than 95% would bail out.
The remaining 5% would bail out once they reach 10M net worth no debt.
CEOs turnover and population is so low because it takes a special kind of monodimensional individual to have the world as your oyster 5-10M in your name and no debt and turn around and say "I'd rather go to work and pretend to be at the helm of an imaginary ship instead"
You are assuming here that boards are always acting in the best interest of the share holders. I bet there are many who do, but there are at least as many that act primarily in the interest of themselves. Board membership is a lucrative job without much supervision, if any at all. So nothing is keeping board members from taking it easy. Why rock the boat if this can make you known as difficult? You might not get that 2nd or 3rd board membership then anymore.
You might remember Credit Suisse, the major bank that had to be taken over recently by UBS to avoid bankruptcy: until then it had gone from scandal to scandal for 15 years in a row. 2 CEO's had been sent away in that period (by shareholders) but the president of the board? He could stay, and with him the rest of the board. His salary? 7 million a year.
The whole purpose of title inflation is to direct money to your friends with plausible deniability.
Now someone will bring up the rare cases where a CEO actually did something. That does not disprove the rule.
Citation needed, etc. I think this is just meant to be funny but if anyone wants it to be more than that I’d need to see some actual proof.
The quote from the article is clearly from someone who has never run a company, of size at least.
People have a very hard time being good CEO's, AI will not make that happen any time soon.
(I'd be cool with CEOs making less than 400x their workers though)
Power is the ability to do more work, faster.
I'm paid well. Do everything I'm told. Top performer. I'm happy though as I'm providing for the family. My accomplishments amount to one day of effort, output, results every month.
This has been happening for years.
I am curious how many of me exist. The Twitter firing suggests that's it's over 50% so it's that's true, what exactly are managers doing?
Most tell me they're in meetings, looking over spreadsheets and meeting with staff. I run my house and meet with my spouse for about twenty minutes a quarter over financials. Everything is automated. Everything is scheduled. We are adults. Labor is divided, assigned, adjusted as we age.
Large companies get in the position of not reducing head count because, yes, they're people with lives and relationships matter.
Solution? I'm thinking more of a State style employment but limited. If hired, you're job is secure for five years. Automate you job entirely? Good on you! Do not apply...... TAKE a new job in the company. You're slow to onboard? Fine. Dead weight? Not renewed.
Take the fear of insecurity off the table. People will behave according to their character.